Inventors list

Assignees list

Classification tree browser

Top 100 Inventors

Top 100 Assignees


MACHINE LEARNING

Subclass of:

706 - Data processing: artificial intelligence

Patent class list (only not empty are listed)

Deeper subclasses:

Class / Patent application numberDescriptionNumber of patent applications / Date published
706013000 Genetic algorithm and genetic programming system 159
Entries
DocumentTitleDate
20080281764Machine Learning System - A method for training a classifier to classify elements of a data set according to a characteristic is described. The data set includes N elements with the elements each characterized by at least one feature. The method includes the steps of forming a first labeled subset of elements from the data set with the elements of the first labeled subset each labeled according to whether the element includes the characteristic, training an algorithmic classifier to classify for the characteristic according to the first labeled subset thereby determining which at least one feature is relevant to classifying for the characteristic; and then querying with the classifier an inverted index, with this inverted index formed over the at least one feature and generated from the data set, thereby generating a ranked set of elements from the data set.11-13-2008
20110178964Recommendation System Using Rough-Set and Multiple Features Mining Integrally and Method Thereof - The present invention solves problems of cold start, first rater, sparsity and scalability for recommendation. A recommendation system according to the present invention finds association rules through data mining. Then, the recommendation system integrates a rough-set algorithm and a statistical analysis prediction for recommendation. The recommendation is dynamically made from a result of the rough-set algorithm and a result of the statistical analysis prediction by setting a standard deviation as a threshold.07-21-2011
20090216693CLASSIFICATION METHOD AND APPARATUS - A method for building a classification model for classifying unclassified documents based on the classification of a plurality of documents which respectively have been classified as belonging to one of a plurality of classes, said documents being digitally represented in a computer, said documents respectively comprising a plurality of terms which respectively comprise one or more symbols of a finite set of symbols, and said method comprising the following steps: representing each of said plurality of documents by a vector of n dimensions, said n dimensions forming a vector spaces whereas the value of each dimension of said vector corresponds to the frequency of occurrence of a certain term in the document corresponding to said vector, so that said n dimensions span up a vector space; representing the classification of said already classified documents into classes by separating said vector space into a plurality of subspaces by one or more hyperplanes, such that each subspace comprises one or more documents as represented by their corresponding vectors in said vector space, so that said each subspace corresponds to a class.08-27-2009
20100088256Method and monitoring system for the rule-based monitoring of a service-oriented architecture - The present invention concerns a method for the rule-based monitoring of a component (C04-08-2010
20090024546SYSTEM, METHOD AND APPARATUS FOR PREDICTIVE MODELING OF SPATIALLY DISTRIBUTED DATA FOR LOCATION BASED COMMERCIAL SERVICES - A computer system implements a method to provide a class membership probability prediction based on collected usage data from a user device of a user. After device usage data, which contains location information, is collected from the user device, the collected usage data is processed to generate a predictive model by utilizing a machine learning algorithm. In response to a user input, a class membership probability estimation is produced by processing the user input through the probability predictive model. The resulted class membership probability estimation can then be used as a prediction of a demographic profile of the user.01-22-2009
20110191276OPEN INFORMATION EXTRACTION FROM THE WEB - To implement open information extraction, a new extraction paradigm has been developed in which a system makes a single data-driven pass over a corpus of text, extracting a large set of relational tuples without requiring any human input. Using training data, a Self-Supervised Learner employs a parser and heuristics to determine criteria that will be used by an extraction classifier (or other ranking model) for evaluating the trustworthiness of candidate tuples that have been extracted from the corpus of text, by applying heuristics to the corpus of text. The classifier retains tuples with a sufficiently high probability of being trustworthy. A redundancy-based assessor assigns a probability to each retained tuple to indicate a likelihood that the retained tuple is an actual instance of a relationship between a plurality of objects comprising the retained tuple. The retained tuples comprise an extraction graph that can be queried for information.08-04-2011
20110191275SYSTEM AND METHOD TO ESTIMATE REGION OF TISSUE ACTIVATION - A computer-implemented method for determining the volume of activation of neural tissue. In one embodiment, the method uses one or more parametric equations that define a volume of activation, wherein the parameters for the one or more parametric equations are given as a function of an input vector that includes stimulation parameters. After receiving input data that includes values for the stimulation parameters and defining the input vector using the input data, the input vector is applied to the function to obtain the parameters for the one or more parametric equations. The parametric equation is solved to obtain a calculated volume of activation.08-04-2011
20100153315BOOSTING ALGORITHM FOR RANKING MODEL ADAPTATION - Model adaptation may be performed to take a general model trained with a set of training data (possibly large), and adapt the model using a set of domain-specific training data (possibly small). The parameters, structure, or configuration of a model trained in one domain (called the background domain) may be adapted to a different domain (called the adaptation domain), for which there may be a limited amount of training data. The adaption may be performed using the Boosting Algorithm to select an optimal basis function that optimizes a measure of error of the model as it is being iteratively refined, i.e., adapted.06-17-2010
20100153318Methods and systems for automatically summarizing semantic properties from documents with freeform textual annotations - Some embodiments are directed to identifying semantic properties of documents using free-text annotations associated with the documents. Semantic properties of documents may be identified by using a model that is trained on a corpus of training documents where one or more of the training documents may include free-text annotations. In some embodiments, the model may identify semantic topics expressed only in free-text annotations or only in the body of a document. The model may applied to identify semantic topics associated with a work document or to summarize the semantic topics present in a plurality of work documents.06-17-2010
20090192956METHOD AND APPARATUS FOR STRUCTURING DOCUMENTS UTILIZING RECOGNITION OF AN ORDERED SEQUENCE OF IDENTIFIERS - A method is provided for operating a computing device to create a document structure model of a computer parsable text document utilizing recognition of at least one ordered sequence of identifiers in the document. The method includes converting a computer parsable text document of any format to an alternative structured language format to form a converted document. The text of the converted document is fragmented into an ordered sequence of text fragments within a text format. The text fragments are enumerated to obtain a sequence of terms. At least one optimal sub-sequence of terms is identified from among the sequence of terms, with an optimal sub-sequence being one or more longest increasing sub-sequence(s). The computer parsable text document is annotated with tags, with the tags including information derived from identification of the optimal sub-sequence(s). The annotated document is displayed on the graphical user interface.07-30-2009
20090192955GRANULAR SUPPORT VECTOR MACHINE WITH RANDOM GRANULARITY - Methods and systems for granular support vector machines. Granular support vector machines can randomly select samples of datapoints and project the samples of datapoints into a randomly selected subspaces to derive granules. A support vector machine can then be used to identify hyperplane classifiers respectively associated with the granules. The hyperplane classifiers can be used on an unknown datapoint to provide a plurality of predictions which can be aggregated to provide a final prediction associated with the datapoint.07-30-2009
20090106174METHODS, SYSTEMS, AND COMPUTER PROGRAM PRODUCTS EXTRACTING NETWORK BEHAVIORAL METRICS AND TRACKING NETWORK BEHAVIORAL CHANGES - A network behavioral metric is extracted from a communication network based on a relevancy of the metric to network behavior by identifying a network metric x that is defined as a random variable that represents a quantitative measure of a network behavior accumulated over a period of time, selecting a network feature, generating a metric disintegration model for the network metric x comprising at least one normal behavior probability distribution function for the metric x for each value of the network feature, respectively, and at least one abnormal behavior probability distribution function for the metric x for each value of the network feature, respectively, increasing a number of the values of the metric x that indicates normal network behavior and/or abnormal network behavior based on the metric disintegration model, and selecting a network metric x as a behavioral metric based on a relevancy η of the network metric x to the network behavior. Embodiments for tracking network behavioral changes are also provided.04-23-2009
20130031038System and Method For Multiclass Discrimination of Neural Response Data - Systems and methods are described herein for analyzing neural response data that can be assigned to multiple classes. The systems and methods begin with a set of training data from which optimal weight factors are derived. The derived weight factors are used in a classifier which is then applied to test data from test subjects. The classifier filters out the effects of less relevant data in the test data and provides a result in the form of probabilities associated with classes for the test data.01-31-2013
20130031037SYSTEM AND METHODOLOGY PROVIDING AUTOMATION SECURITY ANALYSIS AND NETWORK INTRUSION PROTECTION IN AN INDUSTRIAL ENVIRONMENT - The present invention relates to a system and methodology facilitating automation security in a networked-based industrial controller environment. Various components, systems and methodologies are provided to facilitate varying levels of automation security in accordance with security analysis tools, security validation tools and/or security learning systems. The security analysis tool receives abstract factory models or descriptions for input and generates an output that can include security guidelines, components, topologies, procedures, rules, policies, and the like for deployment in an automation security network. The validation tools are operative in the automation security network, wherein the tools perform security checking and/or auditing functions, for example, to determine if security components are in place and/or in suitable working order. The security learning system monitors/learns network traffic patterns during a learning phase, fires alarms or events based upon detected deviations from the learned patterns, and/or causes other automated actions to occur.01-31-2013
20130031034ADAPTIVE RANKING OF NEWS FEED IN SOCIAL NETWORKING SYSTEMS - Machine learning models are used for ranking news feed stories presented to users of a social networking system. The social networking system divides its users into different sets, for example, based on demographic characteristics of the users and generates one model for each set of users. The models are periodically retrained. The news feed ranking model may rank news feeds for a user based on information describing other users connected to the user in the social networking system. Information describing other users connected to the user includes interactions of the other users with objects associated with news feed stories. These interactions include commenting on a news feed story, liking a news feed story, or retrieving information, for example, images, videos associated with a news feed story.01-31-2013
20130031036PARAMETER SETTING APPARATUS, NON-TRANSITORY MEDIUM STORING COMPUTER PROGRAM, AND PARAMETER SETTING METHOD - A parameter setting apparatus for a control parameter for a wireless communication network including a processor, wherein optimizations for optimizing the control parameter are separated into groups which are unrelated to each other, and the processor executes: a first agent program which are assigned to a group-by-group selects an optimization to be activated according to a first value function; and a second agent program which learns a second value function for determining whether an optimization that affects the first value function is to be activated or not and determines whether the optimization is to be activated or not according to the second value function, and, the activation of the optimization by the first agent program is stopped when the second agent program activates the optimization.01-31-2013
20130031035LEARNING ADMISSION POLICY FOR OPTIMIZING QUALITY OF SERVICE OF COMPUTING RESOURCES NETWORKS - A system for learning admission policy for optimizing quality of service of computer resources networks is provided herein. The system includes a statistical data extractor configured to extract historical data of deployment requests issued to an admission unit of a computer resources network. The system further includes a Markov decision process simulator configured to generate a simulation model based on the extracted historical data and resources specifications of the computer resources network, in terms of a Markov decision process. The system further includes a value function generator configured to determine a value function for deployment requests admissions. The system further includes a machine learning unit configured to train a classifier based on the simulation model and the value function, to yield an admission policy usable for processing incoming deployment requests.01-31-2013
20130031033SYSTEM AND METHOD FOR IMPLEMENTING A LEARNING MODEL FOR PREDICTING THE GEOGRAPHIC LOCATION OF AN INTERNET PROTOCOL ADDRESS - A system and method for implementing a learning model for predicting the geographic location of an Internet Protocol (IP) address are disclosed. A particular embodiment of the system and method includes receiving a model to predict a geographic coordinates position of an Internet Protocol (IP) address, the model including one or more parameters and one or more variables associated with coordinates of the IP address and corresponding information associated with the IP address; receiving training data including a plurality of pairs of coordinates of a target IP address and corresponding information associated with the target IP address; determining, by use of a processor, the one or more parameters based on the training data and the model; and returning a result including information indicative of the determined parameters.01-31-2013
20130031032UTILIZATION OF FEATURES EXTRACTED FROM STRUCTURED DOCUMENTS TO IMPROVE SEARCH RELEVANCE - Features automatically extracted from semi-structured web pages are utilized by a search engine to rank documents that include semi-structured web pages. These features include, but are not limited to, a number of reviews, a number of positive reviews, and/or a number of negative reviews from a web page that includes user reviews. These features also include a number of views of a video that is viewable by way of a semi-structured web page. The features also include a number of subscribers to broadcasts of an individual from a social networking web page and a number of contacts of an individual listed on a social networking web page.01-31-2013
20130185232PROBABILISTIC EVENT NETWORKS BASED ON DISTRIBUTED TIME-STAMPED DATA - Described herein are techniques for producing probabilistic event networks (Bayesian network based representation of node dependencies, whereas nodes comprise event occurrences, explicit times of occurrences, and the context of event occurrences) based on distributed time-stamped data. An aspect provides a method for predicting events from event log data via constructing a probabilistic event net and using the probabilistic event net to infer a probabilistic statement regarding a future event using a network inference mechanism. Other embodiments are disclosed.07-18-2013
20110202485PCC/QOS RULE CREATION - Various exemplary embodiments relate to a method and related network node and machine-readable storage medium including one or more of the following: receiving, at the PCRN, the application request message; determining at least one requested service flow from the application request message; for each requested service flow of the at least one requested service flow, generating a new PCC rule based on the application request message; and providing each new PCC rule to a Policy and Charging Enforcement Node (PCEN). Various exemplary embodiments further include an application request message including at least one media component and at least one media subcomponent and the step of for each media subcomponent, determining a requested service flow from the media subcomponent.08-18-2011
20110202484ANALYZING PARALLEL TOPICS FROM CORRELATED DOCUMENTS - Access is obtained to a parallel corpus including a problem corpus and a solution corpus. A first plurality of topics are mined from the problem corpus and a second plurality of topics are mined from the solution corpus. A transition probability from the first plurality of topics to the second plurality of topics is determined, to identify a most appropriate one of the topics from the solution corpus for a given one of the topics from the problem corpus.08-18-2011
20110202488Method And Apparatus For Creating State Estimation Models In Machine Condition Monitoring - In a machine condition monitoring technique, related sensors are grouped together in clusters to improve the performance of state estimation models. To form the clusters, the entire set of sensors is first analyzed using a Gaussian process regression (GPR) to make a prediction of each sensor from the others in the set. A dependency analysis of the GPR then uses thresholds to determine which sensors are related. Related sensors are then placed together in clusters. State estimation models utilizing the clusters of sensors may then be trained.08-18-2011
20110202487STATISTICAL MODEL LEARNING DEVICE, STATISTICAL MODEL LEARNING METHOD, AND PROGRAM - A statistical model learning device is provided to efficiently select data effective in improving the quality of statistical models. A data classification means 08-18-2011
20110202486Healthcare Information Technology System for Predicting Development of Cardiovascular Conditions - Described herein is a framework for predicting development of a cardiovascular condition of interest in a patient. The framework involves determining, based on prior domain knowledge relating to the cardiovascular condition of interest, a risk score as a function of patient data. The patient data may include both genetic data and non-genetic data. In one implementation, the risk score is used to categorize the patient into at least one of multiple risk categories, the multiple risk categories being associated with different strategies to prevent the onset of the cardiovascular condition. The results generated by the framework may be presented to a physician to facilitate interpretation, risk assessment and/or clinical decision support.08-18-2011
20120179635METHOD AND SYSTEM FOR MULTIPLE DATASET GAUSSIAN PROCESS MODELING - A method of computerised data analysis and synthesis is described. First and second datasets of a quantity of interest are stored. A Gaussian process model is generated using the first and second datasets to compute optimized kernel and noise hyperparameters. The Gaussian process model is applied using the stored first and second datasets and hyperparameters to perform Gaussian process regression to compute estimates of unknown values of the quantity of interest. The resulting computed estimates of the quantity of interest result from a non-parametric Gaussian process fusion of the first and second measurement datasets. The first and second datasets may be derived from the same or different measurement sensors. Different sensors may have different noise and/or other characteristics.07-12-2012
20100076912System and Method for Determining a Characteristic of an Individual - A system and method for determining a characteristic of an individual is provided. The method includes determining at least one nonconscious element of an interaction by the individual and correlating the at least one nonconscious element with at least one identifiable demographic characteristic of the individual. The system includes a computerized medium having a human interface system situated to facilitate interaction with the individual and produce a quantity of data corresponding to the interaction. A programmable device is in communication with the computerized medium and is situated to use at least a portion of the quantity of data corresponding to the interaction with the individual to determine at least one nonconscious element of the interaction with the individual. A correlation system is situated to correlate the at least one nonconscious element with at least one identifiable demographic characteristic and output a quantity of resulting information.03-25-2010
20100076911Automated Feature Selection Based on Rankboost for Ranking - A method using a RankBoost-based algorithm to automatically select features for further ranking model training is provided. The method reiteratively applies a set of ranking candidates to a training data set comprising a plurality of ranking objects having a known pairwise ranking order. Each round of iteration applies a weight distribution of ranking object pairs, yields a ranking result by each ranking candidate, identifies a favored ranking candidate for the round based on the ranking results, and updates the weight distribution to be used in next iteration round by increasing weights of ranking object pairs that are poorly ranked by the favored ranking candidate. The method then infers a target feature set from the favored ranking candidates identified in the iterations.03-25-2010
20120246099LEARNING DEVICE, LEARNING METHOD, AND COMPUTER PROGRAM PRODUCT - According to an embodiment, a learning device includes a selecting unit, a learning unit, and an evaluating unit. The selecting unit performs a plurality of selection processes of selecting a plurality of groups including one or more learning samples from a learning sample storage unit, where respective learning samples are classified into any one of a plurality of categories. The learning unit learns a classification metric and obtains a set of a classification metric. The evaluating unit acquires two or more evaluation samples of different categories from an evaluation sample storage unit where respective evaluation samples are classified into any one of a plurality of categories; evaluates the classification metric included in the set of the classification metric using the two or more acquired evaluation samples; acquires a plurality of classification metric corresponding to the evaluation results from the set of the classification metric; and thereby generates an evaluation metric including the plurality of classification metric.09-27-2012
20120246098Role Mining With User Attribution Using Generative Models - Applications of machine learning techniques such as Latent Dirichlet Allocation (LDA) and author-topic models (ATM) to the problems of mining of user roles to specify access control policies from entitlement as well as logs which contain record of the usage of these entitlements are provided. In one aspect, a method for performing role mining given a plurality of users and a plurality of permissions is provided. The method includes the following steps. At least one generative machine learning technique, e.g., LDA, is used to obtain a probability distribution θ for user-to-role assignments and a probability distribution β for role-to-permission assignments. The probability distribution θ for user-to-role assignments and the probability distribution β for role-to-permission assignments are used to produce a final set of roles, including user-to-role assignments and role-to-permission assignments.09-27-2012
20100114803APPARATUS AND METHOD FOR MODELING USER'S SERVICE USE PATTERN - Provided are an apparatus and method for learning and modeling a user's service use pattern. The method includes: collecting information about a service selected by the user and situation information of the user when selecting the service; learning the user's service use pattern based on the collected information; and updating a learning value of a corresponding context-service pair in a user model, which is comprised of context-service pairs, based on the learning result, wherein the situation information of the user includes one or more contexts.05-06-2010
20130085971AUTOMATIC TRACE RETRIEVAL USING SEMANTIC BRIDGE - A method for performing automatic trace retrieval includes receiving a first and second model for a system or service (S04-04-2013
20130080359ASSISTING VEHICLE GUIDANCE OVER TERRAIN - A method and system for assisting with guiding a vehicle over terrain is provided. The method includes training at least one first classifier technique using a first set of terrain classifier training data, such that the at least one first classifier technique is trained to output at least one probability value usable to classify terrain. The first trained classifier technique is then used to generate a second set of terrain classifier training data. A second classifier technique is trained using the output of the at least one first classifier technique, and additional data to output a probability value useable to classify terrain.03-28-2013
20100145895COMPONENT RELIABILITY BUDGETING SYSTEM - A system may include acquisition of a supply voltage information representing past supply voltages supplied to an electrical component, acquisition of a temperature information representing past temperatures of the electrical component, and control of a performance characteristic of the electrical component based on the supply voltage information and the temperature information. Some embodiments may further include determination of a reliability margin based on the supply voltage information, the temperature information, and on a reliability specification of the electrical component, and change of the performance characteristic based on the reliability margin.06-10-2010
20130138587SYSTEM AND METHOD FOR GRAPH PATTERN ANALYSIS - In some example embodiments, a system and method are provided for graph pattern analysis. In example embodiments, pattern data of a primary network that includes data relating to relationships between entities are received. A secondary network based on the pattern data of the primary network is generated by using an algorithm that processes pattern characteristics extracted from the pattern data. The generated secondary network is provided for further analysis.05-30-2013
20130036076METHOD FOR KEYWORD EXTRACTION - Presented is a method of extracting keywords. The method includes obtaining a corpus of documents, determining a first set of words that appear as keywords in a document present in the corpus of documents, determining a second set of words that appear in the corpus of documents but not necessarily appear as keywords in the document, and determining a final set of keywords for the document by combining the first set of words with the second set of words.02-07-2013
20120265718METHOD AND APPARATUS FOR EVOLVING A QUANTUM SYSTEM USING A MIXED INITIAL HAMILTONIAN COMPRISING BOTH DIAGONAL AND OFF-DIAGONAL TERMS - Various adaptations to adiabatic quantum computation and quantum annealing are described. These adaptations generally involve tailoring an initial Hamiltonian so that a local minimum is avoided when a quantum processor is evolved from the initial Hamiltonian to a problem Hamiltonian. The initial Hamiltonian may represent a mixed Hamiltonian that includes both diagonal and off-diagonal terms, where the diagonal terms at least partially define a center point of a first computation space that is at least partially contained within a second computation space. A problem Hamiltonian may be evolved into a low energy state by inhomogeneously inducing disorder in the qubits of the quantum processor. A higher degree of disorder may be induced in a subset of qubits predicted to contribute to a local minimum of the problem Hamiltonian.10-18-2012
20090171866System and method for learning associations between logical objects and determining relevance based upon user activity - User activity streams are used to automatically learn associations between logical objects and form logical groups. Search results are sorted based upon their relevance to current user activity. Combined with a graphical user interface component and object database, the invention automatically retrieves and display groups and objects related to the active object.07-02-2009
20120209795WEB PAGE ANALYSIS SYSTEM FOR COMPUTERIZED DERIVATION OF WEBPAGE AUDIENCE CHARACTERISTICS - A system for generating a demographic profile for a set of at least one webpage, the system comprising a webpage audience information gatherer operative for providing, for at least one webpage, training data including demographic information characterizing an audience of the webpage; a predictor developing system operative to compute at least one content characteristic of said webpage and to develop a prediction process which if applied to said content characteristic would have predicted said training data; and a webpage audience predictor operative, for at least one new webpage, whose audience is unknown, to compute at least one content characteristic of the new webpage and to generate predicted demographic information predicted to characterize said unknown audience of said new webpage by applying said prediction process to said new webpage's content characteristic.08-16-2012
20130041856METHOD FOR DETECTION OF MOVEMENT OF A SPECIFIC TYPE OF OBJECT OR ANIMAL BASED ON RADAR SIGNALS - A method of detecting movement includes using a radar sensor to monitor a space, and receiving an output signal from the radar sensor. A Fourier transform is performed on the output signal to produce a frequency domain signal spectrum. The frequency domain signal spectrum is transformed into an acoustic domain signal. It is decided whether the output signal is indicative of movement of a predetermined object or a non-human animal dependent upon at least one feature of the acoustic domain signal and at least one spectral feature of the signal spectrum.02-14-2013
20100042560CONTEXT AWARE SOLUTION ASSEMBLY IN CONTACT CENTER APPLICATIONS - An apparatus and a method is provided for receiving help requests to solve a problem on a computer, generating a core problem description and retrieving at least one of contextual or environmental parameters associated with the computer. The method also includes assembling a formalized problem description. The method further includes obtaining previously stored formalized solution steps associated with the problem from a database and building a customized solution including context aware solution records that are tagged with at least one of contextual or environmental dependencies. The method also includes transmitting the customized solution to the computer for execution and monitoring the execution of the customized solution.02-18-2010
20100042563SYSTEMS AND METHODS OF DISCOVERING MIXTURES OF MODELS WITHIN DATA AND PROBABILISTIC CLASSIFICATION OF DATA ACCORDING TO THE MODEL MIXTURE - Discovering mixtures of models includes: initiating learning algorithms, determining, data sets including a cluster of points in a first region of a domain and a set of points distributed near a first line extending across the domain; inferencing parameters from the cluster and the set of points; creating a description of the cluster of points in the first region of the domain and computing approximations of a first learned mixture model and a second learned mixture model; determining a first and second probability, generating a confidence rating that each point of the cluster of points in the first region of the domain corresponds to the first learned mixture model and generating a confidence rating that each point of the set of points distributed near the first line correspond to the second learned mixture model, thus causing determinations of behavior of a system described by the learned mixture models.02-18-2010
20100042562METHOD OF DOWNLOADING USAGE PARAMETERS INTO AN APPARATUS, AND APPARATUS FOR IMPLEMENTING THE INVENTION - Method of downloading usage parameters into an apparatus, and apparatus for implementing the invention After a first start-up, a first appliance performs a self-learning step for generating usage parameters. These parameters are elaborated on subsequent start-ups. When these parameters are optimized, the first appliance transmits them to another appliance which requests them. This second appliance uses the parameters of the first as optimized parameters. In this way, the second appliance limits the duration of the self-learning step and the use of non-optimal parameters. According to a refinement, the optimal parameters are centralized on a server which transmits them to a plurality of second appliances using a transmission network.02-18-2010
20100042561METHODS AND SYSTEMS FOR COST-SENSITIVE BOOSTING - Multi-class cost-sensitive boosting based on gradient boosting with “p-norm” cost functionals” uses iterative example weighting schemes derived with respect to cost functionals, and a binary classification algorithm. Weighted sampling is iteratively applied from an expanded data set obtained by enhancing each example in the original data set with as many data points as there are possible labels for any single instance, and where each non-optimally labeled example is given the weight equaling a half times the original misclassification cost for the labeled example times the p−1 norm of the average prediction of the current hypotheses. Each optimally labeled example is given the weight equaling the sum of the weights for all the non-optimally labeled examples for the same instance. Component classification algorithm is executed on a modified binary classification problem. A classifier hypothesis is output, which is the average of all the hypotheses output in the respective iterations.02-18-2010
20100161525ANALYZING A TARGET ELECTROMAGNETIC SIGNAL RADIATING FROM A COMPUTER SYSTEM - One embodiment of the present invention provides a system that characterizes a computer system parameter by analyzing a target electromagnetic signal radiating from the computer system. First, the system monitors the target electromagnetic signal using a first directional antenna located outside of the computer system, wherein the first directional antenna is directed at a location inside the computer system. The system also monitors the target electromagnetic signal using a second directional antenna located outside of the computer system, wherein a receiving axis of the second antenna is oriented non-parallel to a receiving axis of the first antenna, and wherein the second directional antenna is directed at the location inside the computer system. Next, the system characterizes the computer system parameter based on the target electromagnetic signal received from the first antenna and the target electromagnetic signal received from the second antenna. Then, the system generates a request for an action based on the characterization of the computer system.06-24-2010
20100161523GENERATION AND USE OF SPECIFIC PROBABILITY TABLES FOR ARITHMETIC CODING IN DATA COMPRESSION SYSTEMS - In one embodiment, when executing data compression or decompression for a data set, a particular compression category of the data set is determined, and a corresponding probability table specific to the particular compression category of the data set is accessed. Then, one of either arithmetic coding (e.g., an encoder device) or decoding (e.g., a decoder device) may be performed on the data set based on the specific probability table. Specifically, in one or more other embodiments, techniques may statistically generate probability tables specific to particular compression categories.06-24-2010
20100161526Ranking With Learned Rules - Systems, methods and computer program products for the ranking of a target data set based on learned rules are disclosed. One embodiment is a method that includes generating a learned rule set from a training data record set, creating at least one prototype for each rule in the learned rule set to generate a prototype set, and ranking the target data record set using learned rule set and the prototype set. The generating of a learned rule set includes dividing the training data record set to a positive class and a negative class, and deriving the learned rule set for the positive class. Learning of rules includes deriving the most general projected rules with respect to remaining training data and then refining those rules, eventually selecting the best rules using an F-measure.06-24-2010
20090157574METHOD AND APPARATUS FOR ANALYZING WEB SERVER LOG BY INTRUSION DETECTION SYSTEM - Provided is hacking prevention technology, and more particularly, a method and apparatus for automatically analyzing log information of a web server for which intrusion is attempted from an outside source.06-18-2009
20090157572STACKED GENERALIZATION LEARNING FOR DOCUMENT ANNOTATION - A document annotation method includes modeling data elements of an input document and dependencies between the data elements as a dependency network. Static features of at least some of the data elements are defined, each expressing a relationship between a characteristic of the data element and its label. Dynamic features are defined which define links between an element and labels of the element and of a second element. Parameters of a collective probabilistic model for the document are learned, each expressing a conditional probability that a first data element should be labeled with information derived from a label of a neighbor data element linked to the first data element by a dynamic feature. The learning includes decomposing a globally trained model into a set of local learning models. The local learning models each employ static features to generate estimations of the neighbor element labels for at least one of the data elements.06-18-2009
20090157571METHOD AND APPARATUS FOR MODEL-SHARED SUBSPACE BOOSTING FOR MULTI-LABEL CLASSIFICATION - A computer program product includes machine readable instructions for managing data items, the instructions stored on machine readable media, the product including instructions for: initializing a plurality of base models; minimizing a joint loss function to select models from the plurality for a plurality of labels associated with the data items; and at least one of sharing and combining the selected base models to formulate a composite classifier for each label. A computer system and additional computer program product are provided.06-18-2009
20130046715METHOD TO DETERMINE AN ARTIFICIAL LIMB MOVEMENT FROM AN ELECTROENCEPHALOGRAPHIC SIGNAL - The present invention is related to a method to determine an artificial limb movement comprising the steps of: providing an EEG input training dataset; providing an output prosthetic limb movement training dataset corresponding to said EEG input training dataset; providing a dynamic recurrent neural network (DRNN) comprising a convergence acceleration algorithm; training said DRNN with said input and output datasets to define synaptic weights W02-21-2013
20130046714Evaluating the health status of a system - A method and apparatus for determining a health of the system. Groups of vibration data are identified for the system. A group of vibration data in the groups of vibration data comprises data for vibrations of the system at different frequencies over time. The groups of vibration data for the system are stored in a number of associative memories in a computer system. The health of the system is identified based on the groups of vibration data in the number of associative memories.02-21-2013
20090125462Method and system using keyword vectors and associated metrics for learning and prediction of user correlation of targeted content messages in a mobile environment - Methods and systems for determining a suitability for a mobile client to display information are disclosed. A particular exemplary method includes receiving a plurality of sets of one or more first keywords on a mobile client, each set of first keywords associated with one or more respective first messages, monitoring user interaction of the respective first messages on the mobile client, performing learning operations on the mobile client with the first keywords based on monitored user interaction to estimate a set of keyword interest weights, receiving a set of target keywords associated with a target message, and displaying the target message on the mobile client based on the estimated keyword interest weights.05-14-2009
20090125461Multi-Label Active Learning - Multi-label active learning may entail training a classifier with a set of training samples having multiple labels per sample. In an example embodiment, a method includes accepting a set of training samples, with the set of training samples having multiple respective samples that are each respectively associated with multiple labels. The set of training samples is analyzed to select a sample-label pair responsive to at least one error parameter. The selected sample-label pair is then submitted to an oracle for labeling.05-14-2009
20100094783Method and System for Classifying Data in System with Limited Memory - Embodiments of the invention describe a method for classifying data in a system with limited memory. The method applies exemplar learning (EL) procedures to a training data set to produce an exemplar data set adapted to the size of the memory. The EL procedure is selected form a group consisting of an entropy based exemplar learning (EBEL) procedure and an advanced broadband enabled learning (ABEL) procedure. The exemplar data set is used to classify acquired by the system data.04-15-2010
20090012921METHOD FOR IDENTIFYING A PERSON'S POSTURE - A classification method including first classifying an event of any kind by first rules, and then second classifying events, not identified by the first classification, by a learning base reinforced with all the events identified by the first classification. The method is adaptive if the second classification rules are amended according to new examples that were able to be determined by the first rules.01-08-2009
20100070435Computationally Efficient Probabilistic Linear Regression - A computationally efficient method of performing probabilistic linear regression is described. In an embodiment, the method involves adding a white noise term to a weighted linear sum of basis functions and then normalizing the combination. This generates a linear model comprising a set of sparse, normalized basis functions and a modulated noise term. When using the linear model to perform linear regression, the modulated noise term increases the variance associated with output values which are distant from any data points.03-18-2010
20090043717METHOD AND A SYSTEM FOR SOLVING DIFFICULT LEARNING PROBLEMS USING CASCADES OF WEAK LEARNERS - A method and a system for designing a learning system (02-12-2009
20090043716DATA CLASSIFICATION METHOD AND APPARATUS - A data classification apparatus for classifying plural input data into plural categories, in which the apparatus includes a prototype select unit for selecting the prototype of the category nearest to the input data that has been read, a prototype evaluation unit for evaluating whether the selected prototype is proper, a prototype addition unit for adding a prototype in the case where the selected prototype is not proper and an internal data correcting unit for correcting at least one of the prototype and an area determining parameter specifying the size of the category area for each category in the case where the selected prototype is proper. The size of the category area can be set for each category, and therefore, the data can be properly classified and the judgment accuracy is improved in an application to fault detection and fault diagnosis.02-12-2009
20090043715Method to Continuously Diagnose and Model Changes of Real-Valued Streaming Variables - The method trains an inductive model to output multiple models from the inductive model and trains an error correlation model to estimate an average output of predictions made by the multiple models. Then the method can determine an error estimation of each of the multiple models using the error correlation model.02-12-2009
20110010321MARKOVIAN-SEQUENCE GENERATOR AND NEW METHODS OF GENERATING MARKOVIAN SEQUENCES - A new type of Markovian sequence generator and generation method generates a Markovian sequence having controllable properties, notably properties that satisfy at least one control criterion which is a computable requirement holding on items in the sequence. The Markovian sequence is generated chunkwise, each chunk containing a plurality of items in the sequence. During generation of each chunk a search is performed in the space of Markovian sequences to find a chunk-sized series of items which enables the control criterion to be satisfied. The search can be performed using a generate and test approach in which chunk-sized Markovian sequences are generated then tested for compliance with the requirement(s) of the control criteria. Alternatively, the search can be performed by formulating the sequence-generation task as a constraint satisfaction problem, with one or more constraints ensuring that the generated sequence is Markovian and one or more constraints enforcing the requirement(s) of the control criteria. The sequence generator can be used in an interactive system where a user specifies the control criterion via an inputting device (01-13-2011
20110010319CORRESPONDENCE LEARNING APPARATUS AND METHOD AND CORRESPONDENCE LEARNING PROGRAM, ANNOTATION APPARATUS AND METHOD AND ANNOTATION PROGRAM, AND RETRIEVAL APPARATUS AND METHOD AND RETRIEVAL PROGRAM - An image data processing system has a learning storage apparatus that stores projection matrixes obtained by canonical correlation analysis so as to derive, based on at least one of an image feature and a word feature, a latent variable as an abstract concept used for associating an image with a word corresponding thereto and that further stores information required for obtaining the latent variable acquired by use of the projection matrixes, a probability of occurrence of an arbitrary image feature from a certain latent variable and a probability of occurrence of an arbitrary word feature from a certain latent variable. In this way, a probability of the image feature and word feature being simultaneously outputted can be easily and quickly determined, thereby executing a high-speed annotation or retrieval with high precision.01-13-2011
20090327176SYSTEM AND METHOD FOR LEARNING - A method of learning discriminant function for predicting label information by using computer includes: receiving training data including attribute data and label information, to create an initial prediction model based on the attribute data and the label information; calculating, based on the initial prediction model used as a discriminant function, a gradient of a loss function, which is differentiable with respect to the discriminant function and satisfies a monotonous convex function, from the discriminant function and the label information; creating a prediction model from the attribute data and the gradient while assuming that the gradient is label information of each sample of the training data; and updating the discriminant function based on the created prediction model.12-31-2009
20090307162METHOD AND APPARATUS FOR AUTOMATED ASSISTANCE WITH TASK MANAGEMENT - The present invention relates to a method and apparatus for assisting with automated task management. In one embodiment, an apparatus for assisting a user in the execution of a task, where the task includes one or more workflows required to accomplish a goal defined by the user, includes a task learner for creating new workflows from user demonstrations, a workflow tracker for identifying and tracking the progress of a current workflow executing on a machine used by the user, a task assistance processor coupled to the workflow tracker, for generating a suggestion based on the progress of the current workflow, and a task executor coupled to the task assistance processor, for manipulating an application on the machine used by the user to carry out the suggestion.12-10-2009
20090307161SYSTEM AND METHOD TO LEARN AND DEPLOY AN OPTIMAL USER EXPERIENCE IN AN ONLINE SYSTEM - Methods and systems to learn an optimal user experience. The system receives a request over a network from a user. The request includes context information. The system identifies a response to the request is to be utilized to learn whether a first interface component included in a first plurality of interface components is an optimal choice for a first decision. The response includes an interface. The interface includes the first interface component. The system identifies the response to the request is to be utilized based on the context information. Finally, the system communicates the response over the network to the user.12-10-2009
20090307160PARALLEL GENERATION OF A BAYESIAN NETWORK - A method for generating a Bayesian network in a parallel manner is based on an initial model having a plurality of nodes. Each node corresponds to a variable of a data set and has a local distribution associated therewith. The method includes assigning a plurality of subsets of the nodes to a respective plurality of constructors. The plurality of constructors is operated in a parallel manner to identify edges to add between nodes in the initial model. The identified edges are added to the initial model to generate the Bayesian network. The edges indicate dependency between nodes connected by the edges.12-10-2009
20120191634STORAGE POLICY EVALUATION IN A COMPUTING ENVIRONMENT - Systems and methods for generating a storage policy for a storage system are provided. The method comprises receiving a target function applicable to a storage system having one or more data storage mediums, wherein the target function represents values for storage parameters associated with productivity or loss tolerance in the storage system; implementing one or more simulation rules according to the received target function; generating one or more storage operation requests to access data on said one or more data storage mediums based on said one or more simulation rules; submitting said one or more storage operation requests to the storage system for processing; analyzing simulation results obtained for the storage system, in response to the storage system processing said one or more storage operation requests; and generating one or more storage policies, by a machine learning entity, in response to analyzing the simulation results.07-26-2012
20120191632SYSTEM AND METHODS FOR FINDING HIDDEN TOPICS OF DOCUMENTS AND PREFERENCE RANKING DOCUMENTS - Systems and methods are disclosed to perform preference learning on a set of documents includes receiving raw input features from the set of documents stored on a data storage device; generating polynomial combinations from the raw input features; generating one or more parameters; applying the parameters to one or more classifiers to generate outputs; determining a loss function and parameter gradients and updating parameters determining one or more sparse regularizing terms and updating the parameters; and expressing that one document is preferred over another in a search query and retrieving one or more documents responsive to the search query.07-26-2012
20120191631Dynamic Predictive Modeling Platform - Methods, systems, and apparatus, including computer programs encoded on one or more computer storage devices, for training and retraining predictive models. A series of training data sets are received and added to a training data queue. In response to a first condition being satisfied, multiple retrained predictive models are generated using the training data queue, multiple updateable trained predictive models obtained from a repository of trained predictive models, and multiple training functions. In response to a second condition being satisfied, multiple new trained predictive models are generated using the training data queue, at least some training data stored in a training data repository and training functions. The new trained predictive models include static trained predictive models and updateable trained predictive models. The repository of trained predictive models is updated with at least some of the retrained predictive models and new trained predictive models.07-26-2012
20110016065EFFICIENT ALGORITHM FOR PAIRWISE PREFERENCE LEARNING - In one embodiment, training a ranking model comprises: accessing the ranking model and an objective function of the ranking model; accessing one or more preference pairs of objects, wherein for each of the preference pairs of objects comprising a first object and a second object, there is a preference between the first object and the second object with respect to the particular reference, and the first object and the second object each has a feature vector comprising one or more feature values; and training the ranking model by minimizing the objective function using the preference pairs of objects, wherein for each of the preference pairs of objects, a difference between the first feature vector of the first object and the second feature vector of the second object is not calculated.01-20-2011
20130073489HYBRID INTERIOR-POINT ALTERNATING DIRECTIONS ALGORITHM FOR SUPPORT VECTOR MACHINES AND FEATURE SELECTION - A method for training a classifier for selecting features in sparse data sets with high feature dimensionality includes providing a set of data items x and labels y, minimizing a functional of the data items x and associated labels y03-21-2013
20130073488METRICS MONITORING AND FINANCIAL VALIDATION SYSTEM (M2FVS) FOR TRACKING PERFORMANCE OF CAPITAL, OPERATIONS, AND MAINTENANCE INVESTMENTS TO AN INFRASTRUCTURE - Techniques for evaluating the accuracy of a predicted effectiveness of an improvement to an infrastructure include collecting data, representative of at least one pre-defined metric, from the infrastructure during first and second time periods corresponding to before and after a change has been implemented, respectively. A machine learning system can receive compiled data representative of the first time period and generate corresponding machine learning data. A machine learning results evaluator can empirically analyze the generated machine learning data. An implementer can implement the change to the infrastructure based at least in part on the data from a machine learning data outputer. A system performance improvement evaluator can compare the compiled data representative of the first time period to that of the second time period to determine a difference, if any, and compare the difference, if any, to a prediction based on the generated machine learning data.03-21-2013
20130073487METHOD AND APPARATUS FOR UTILIZING USER FEEDBACK TO IMPROVE SIGNIFIER MAPPING - Embodiments disclosed herein may relate to processing a user input comprising a resource identity signifier for a target resource with reference to a heuristic knowledge base utilizing a processor of a computing platform to determine a possible target resource. Embodiments may further relate to learning a social usage of the resource identity signifier from feedback gathered from a plurality of users based at least in part on previous social usage of the resource identity signifier, and may also relate to transmitting a resource locator corresponding to the determined possible target resource to a user computing platform at least in part in response to a determination of the possible target resource having a degree of confidence exceeding a selected threshold.03-21-2013
20130073486SYSTEMS AND METHODS FOR ANALYSIS OF NETWORK EQUIPMENT COMMAND LINE INTERFACE (CLI) AND RUNTIME MANAGEMENT OF USER INTERFACE (UI) GENERATION FOR SAME - Systems and methods are disclosed that may be implemented for network management system (NMS) configuration management support for network devices using a learning and natural language processing application to capture the usage and behavior of the Command Line Interface (CLI) of a network device with the aid of a CLI knowledge model, which in one example may be ontology-based.03-21-2013
20130073485METHOD AND APPARATUS FOR MANAGING RECOMMENDATION MODELS - A platform for managing recommendation models is described. The platform processes and/or facilitates a processing of at least one user identification characteristic associated with at least one device to determine a user identity. The platform further determines at least one communication account active at the at least one device. The platform also causes, at least in part, an association of one or more recommendations models with the user identity, the at least one communication account, the at least one device, or a combination thereof.03-21-2013
20090094176PANDEMIC REMOTE ACCESS DESIGN - In one example embodiment, a system and method is illustrated that includes receiving user count information that includes a user count value and an address identifier. Further, an operation is executed that includes using the user count information to determine whether a limit variable has been exceeded. An operation is executed that removes a member, identified by the address identifier, from a load balancing pool, where the limit variable has been exceed by the user count information. A further operation is shown that includes introducing a device into the load balancing pool, where the user count information is less than or equal to the difference between the limit variable value and a buffer variable.04-09-2009
20130060723METHOD AND SYSTEM FOR A SMART AGENT FOR INFORMATION MANAGEMENT WITH FEED AGGREGATION - The present document describes a system and method for managing electronic information. The system may include an input adapted to receive a search topic from a remote user device over a communication network, and a smart agent module configured to perform a semantic analysis on the topic to generate a set of filter parameters. The system may then to collect electronic information about the search topic from at least one remote electronic information source over the communication network, and filter the collected information using the set of filter parameters to produce a stream of filtered information. The filtered information may then be packaged in discrete information containers and sent to the user device for display.03-07-2013
20110066577Machine Learning Using Relational Databases - Machine learning using relational databases is described. In an embodiment a model of a probabilistic relational database is formed by augmenting relation schemas of a relational database with probabilistic attributes. In an example, the model comprises constraints introduced by linking the probabilistic attributes using factor statements. For example, a compiler translates the model into a factor graph data structure which may be passed to an inference engine to carry out machine learning. For example, this enables machine learning to be integrated with the data and it is not necessary to pre-process or reformat large scale data sets for a particular problem domain. In an embodiment a machine learning system for estimating skills of players in an online gaming environment is provided. In another example, a machine learning system for data mining of medical data is provided. In some examples, missing attribute values are filled using machine learning results.03-17-2011
20090271340Method for the computer-aided learning of a control or adjustment of a technical system - A method for the computer-aided learning of a control of a technical system is provided. An operation of the technical system is characterized by states which the technical system can assume during operation. Actions are executed during the operation and convert a relevant state into a subsequent state. The method is characterized in that, when learning the control, suitable consideration is given to the statistical uncertainty of the training data. This is achieved in that the statistical uncertainty of a quality function which models an optimal operation of the technical system is specified by an uncertainty propagation and is incorporated into an action selection rule when learning. By a correspondingly selectable certainty parameter, the learning method can be adapted to different application scenarios which vary in statistical requirements. The method can be used for learning the control of an operation of a turbine, in particular a gas turbine.10-29-2009
20090271339Hierarchical Recognition Through Semantic Embedding - Computer-implemented systems and methods, including servers, perform structure-based recognition processes that include matching and classification. Preprocessing subsystems and sub-methods embed a set of classes on which a loss function is defined into a semantic space and learn an input mapping between an input space and the semantic space. Recognition subsystems and methods accept a test object, representable in the input space, and apply the input mapping to the test object as part of a recognition process.10-29-2009
20090271338SCALABLE FEATURE SELECTION FOR MULTI-CLASS PROBLEMS - In a feature filtering approach, a set of relevant features and a set of training objects classified respective to a set of classes are provided. A candidate feature and a second feature are selected from the set of relevant features. An approximate Markov blanket criterion is computed that is indicative of whether the candidate feature is redundant in view of the second feature. The approximate Markov blanket criterion includes at least one dependency on less than the entire set of classes. An optimized set of relevant features is defined, consisting of a sub-set of the set of relevant features from which features indicated as redundant by the selecting and computing are removed.10-29-2009
20090089226VISUALIZATION OF NON-TIME SERIES EVENTS - Systems and methods that displays available relationships between internal and external data streams. A coordination component can collect and analyze both the “internal” data stream(s) and the “external” data stream(s) simultaneously, and a visualization component can present a form of a visual cue, on a collection of history data and network data. Accordingly, instead of merely storing data values as function of time, other non-time series correlation states can be employed dynamically to represent data to the user.04-02-2009
20120226643EMPIRICAL DATA MODELING - Methods, apparatuses and systems directed to pattern identification and pattern recognition. In some particular implementations, the invention provides a flexible pattern recognition platform including pattern recognition engines that can be dynamically adjusted to implement specific pattern recognition configurations for individual pattern recognition applications. In some implementations, the present invention also provides for a partition configuration where knowledge elements can be grouped and pattern recognition operations can be individually configured and arranged to allow for multi-level pattern recognition schemes.09-06-2012
20120226642METHOD AND APPARATUS FOR CONSIDERING MULTI-USER PREFERENCE BASED ON MULTI-USER-CRITERIA GROUP - A method and apparatus for decision making considering a multi-user preference based on a multi-user-criterion group are provided. The method includes determining user information using ontology, determining an appointed area and an appointed category based on the user information, determining appointed candidate places belonging to the appointed area and appointed category, and determining a final appointed place among the appointed candidate places based on a user preference.09-06-2012
20120226641TRAINING A SEARCH QUERY INTENT CLASSIFIER USING WIKI ARTICLE TITLES AND A SEARCH CLICK LOG - Techniques are described herein for training a search query intent classifier using wiki article titles and a search click log. Titles of wiki articles that correspond to links that are associated with a specified wiki article and/or titles of wiki articles that are included in a category that includes the specified wiki article are extracted and included with the title of the specified wiki article in an initial set. Each title in the initial set is correlated with respective clicked URI(s) using a search click log. The initial set is expanded to include search terms that are correlated to the clicked URIs based on the search click log to provide an expanded set. The search query intent classifier is trained to classify search queries with respect to a query intent that is associated with the title of the specified wiki article based on the expanded set.09-06-2012
20120226640Behavior and information model to yield more accurate probability of successful outcome - A report indicating a user-reported probability of a successful outcome is received. A behavior and information model is estimated based on the report. The behavior and information model includes a behavior model component having a bias parameter and a consistency parameter. The behavior and information model includes an information model component having a first user-believed probability of a successful outcome and a second user-believed probability of a successful outcome. The behavior and information model is used to yield a model-determined probability of a successful outcome that more accurately reflects a probability of a successful outcome than the user-reported probability of a successful outcome does.09-06-2012
20120226639Systems and Methods for Processing Machine Learning Algorithms in a MapReduce Environment - Systems and methods for processing Machine Learning (ML) algorithms in a MapReduce environment are described. In one embodiment of a method, the method includes receiving a ML algorithm to be executed in the MapReduce environment. The method further includes parsing the ML algorithm into a plurality of statement blocks in a sequence, wherein each statement block comprises a plurality of basic operations (hops). The method also includes automatically determining an execution plan for each statement block, wherein at least one of the execution plans comprises one or more low-level operations (lops). The method further includes implementing the execution plans in the sequence of the plurality of the statement blocks.09-06-2012
20130066816INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD AND PROGRAM - Provided is an information processing apparatus including a learning part performing learning of a model of an environment in which an agent performs action, using an observed value observed in the agent when the agent capable of action performs action, an action determining part determining action to be performed by the agent, based on the model, and a user instruction output part outputting instruction information representing an instruction from a user according to the instruction from the user, wherein the action determining part determines the action performed by the agent according to the instruction information when there is an instruction from the user.03-14-2013
20130066818Automatic Crowd Sourcing for Machine Learning in Information Extraction - A method for enabling machine learning from unstructured documents is described. The method comprises analyzing at an electronic device, one or more structured databases, thereby providing a mapping between a plurality of referenced character strings and a corresponding plurality of type labels; providing, at the electronic device, a first unstructured document comprising a plurality of unstructured character strings; analyzing the first unstructured document to identify a first character string of the plurality of unstructured character strings which is associated with a first referenced character string of the plurality of referenced character strings; associating, within the first unstructured document, a first type label which is mapped to the first referenced character string to the first character string; and determining a training set for machine learning from the first unstructured document comprising the association to the first type label.03-14-2013
20130066817INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD AND PROGRAM - There is provided an information processing apparatus including an information amount gain calculation unit that, on the assumption that a factor that could cause a state transition in a state transition model occurs and the state transition takes place as a result of an occurrence of the factor, determines an information amount gain, which is an information amount obtained by causing the factor to occur regarding a causal relationship between the factor and the state transition and observing a result thereof, an evaluation value calculation unit that determines an evaluation value to evaluate effectiveness of causing each of the factors to occur based on the information amount gain to acquire the causal relationship between the factor and the state transition, and a decision unit that decides the factor to be occurred based on the evaluation value.03-14-2013
20130066815SYSTEM AND METHOD FOR MOBILE CONTEXT DETERMINATION - Methods and a system for mobile device activity classification or context determination. The device compresses and sends sensor data to a remote server together with a selected activity label during a training phase. The remote server receives labeled sensor data from a number of devices and generates a classification model. The model may be reduced to a subspace that represents the dominant model parameters. The subspace data structure, which may be a small matrix, is transmitted to the mobile device. The mobile device uses the subspace data structure to classify device activity as indicated by the device sensors. In one example, the sensor data is projected onto the subspace matrix, which results in estimates of state probabilities for the various predefined states, the dominant one of which is selected as the current state, or estimated state.03-14-2013
20130066814System and Method for Automated Classification of Web pages and Domains - Representative sample pages from websites accessible to Internet users are manually selected and classified into pre-defined categories based on page content to create a training set as an input to a classifier. An automated analysis is performed to identify a list of catchwords comprising the most frequently referenced words, tags, and/or links from the classified samples in each category in the training set. A data mining tool generates unique sets of distinctive catchwords and/or distinctive combinations of catchwords that have a high probability of appearing only in a single one of the pre-defined content categories. The classifier utilizes the sets of distinctive catchwords/combinations to classify new pages into one or more of the pre-defined content categories.03-14-2013
20090248595NAME VERIFICATION USING MACHINE LEARNING - Computer-enabled methods, apparatus, and computer-readable media are provided for verifying that a given network name, such as a URL, is an official, e.g., registered, approved, or otherwise officially recognized, network name that refers to or identifies a principal, such as a business. These techniques involve receiving a principal name and a given network name, receiving at least one feature attribute from at least one database of feature attributes, wherein the at least one feature attribute comprises a characteristic of the principal name or a characteristic of the network name, and invoking a logistic regression method to generate a probability, based upon the at least one feature attribute, that the given network name is an official network name for the principal name. The logistic regression method may include a gradient boosting tree model that generates the probability based upon the at least one feature attribute.10-01-2009
20110022552Systems and Methods for Implementing a Machine-Learning Agent to Retrieve Information in Response to a Message - Mixed-initiative message-augmenting agent systems and methods that provide users with tools that allow them to respond to messages, such as email messages, containing requests for information or otherwise requiring responses that require information that needs to be retrieved from one or more data sources. The systems and methods allow users to train machine-learning agents how to retrieve and present information in responses to like messages so that the machine-learning agents can eventually automatedly generate responses with minimal involvement by the users. Embodiments of the systems and methods allow users to build message-augmenting forms containing the desired information for responding to messages and to demonstrate to the machine-learning agents where to retrieve pertinent information for populating the forms. Embodiments of the systems and methods allow users to modify and repair automatically generated forms to continually improve the knowledge of the machine-learning agents.01-27-2011
20120117009CONSTRUCTING A BAYESIAN NETWORK BASED ON RECEIVED EVENTS ASSOCIATED WITH NETWORK ENTITIES - Records of events associated with network entities in a network environment are received, where the network entities are selected from hardware entities, software entities, and combinations of hardware and software entities. The records of the events are identified to identify relationships between events associated with different ones of the network entities, where the records of the events identify corresponding network entities impacted by the events. A Bayesian network is constructed based on the analyzing, wherein the constructed Bayesian network is able to make predictions regarding relationships between events associated with the network elements.05-10-2012
20120117008Parallel Processing Of Data Sets - Systems, methods, and devices are described for implementing learning algorithms on data sets. A data set may be partitioned into a plurality of data partitions that may be distributed to two or more processors, such as a graphics processing unit. The data partitions may be processed in parallel by each of the processors to determine local counts associated with the data partitions. The local counts may then be aggregated to form a global count that reflects the local counts for the data set. The partitioning may be performed by a data partition algorithm and the processing and the aggregating may be performed by a parallel collapsed Gibbs sampling (CGS) algorithm and/or a parallel collapsed variational Bayesian (CVB) algorithm. In addition, the CGS and/or the CVB algorithms may be associated with the data partition algorithm and may be parallelized to train a latent Dirichlet allocation model.05-10-2012
20120117007Systems and Methods to Facilitate Local Searches via Location Disambiguation - Systems and methods use machine learning techniques to resolve location ambiguity in search queries. In one aspect, a dataset generator generates a training dataset using query logs of a search engine. A training engine applies a machine learning technique to the training dataset to generate a location disambiguation model. A location disambiguation engine uses the location disambiguation model to resolve location ambiguity in subsequent search queries.05-10-2012
20120117006METHOD AND APPARATUS FOR BUILDING A USER BEHAVIOR MODEL - An apparatus may include a monitoring module configured to monitor user interactions by a user with applications. A contextual characteristics determiner may determine one or more contextual characteristics relating to the user interactions, and the contextual characteristics may be categorized based on an ontology model. Thereby, a data model builder may build a user behavior model for the user based at least in part on the user interactions and the contextual characteristics. The apparatus may provide for private storage of the user behavior module. A recommendation module may issue a recommendation, which may be mapped to one of the applications, based at least in part on the user behavior model. The recommendation may be issued in response to a query from a query module. The query may include current contextual characteristics of the user and/or the apparatus.05-10-2012
20120197828Energy Saving Control for Data Center - A data center includes at least one rack containing electronic devices, a data center air conditioning system (DCAC), and an environmental parameter monitoring system. At least one set of eligible environmental parameters is determined that satisfies the cooling demand of the at least one rack containing electronic devices. According to the at least one set of eligible environmental parameters and corresponding relationships between sets of setting parameters of the DCAC and corresponding sets of environmental parameters determined by an artificial neural network, plural sets of setting parameters of the DCAC are determined. A power consumption of the DCAC to which each set of setting parameters in the plural sets of setting parameters corresponds is obtained. A set of setting parameters for which the corresponding power consumption satisfies a predetermined condition for energy saving is selected and us to set the DCAC.08-02-2012
20120197827INFORMATION MATCHING APPARATUS, METHOD OF MATCHING INFORMATION, AND COMPUTER READABLE STORAGE MEDIUM HAVING STORED INFORMATION MATCHING PROGRAM - The information matching apparatus includes: a training data setting unit that sets supervised data in a machine learning device of supervised learning that learns judgment criteria used for a judgment of identicalness, similarity, and relevance between a plurality of records by matching the records configured by sets of values corresponding to items; a check point setting unit that sets a check point configured by one set of two records used for evaluating the set supervised data; and a learning result evaluation unit, for the set check point, acquires a change between a judgment result using judgment criteria derived as a result of learning based on set first supervised data and a judgment result using judgment criteria derived as a result of learning based on set second supervised data set and evaluates the supervised data based on the acquired change.08-02-2012
20090254497METHOD AND SYSTEM FOR DYNAMIC PERFORMANCE MODELING OF COMPUTER APPLICATION SERVICES - A generic queueing network model of a Web services environment is introduced. The behavior of a service is abstracted in three phases: serial, parallel and dormant, thus yielding a Serial Parallel Queueing Network (SPQN) model with a small number of parameters. A method is provided for estimated the parameters of the model that is based on stochastic approximation techniques for solving stochastic optimization problems. The parameter estimation method is shown to perform well in a noisy environment, where performance data is obtained through measurements or using approximate model simulations.10-08-2009
20090240636Method, computer program with program code means and computer program product for analyzing variables influencing a combustion process in a combustion chamber, using a trainable statistical model - The invention relates to sensitivity analysis of variables influencing a combustion process. A trainable, statistical model is trained in such a way that it describes the combustion process in the combustion chamber. The trained statistical model is used to determine the influence of the variables on said combustion process in the combustion chamber.09-24-2009
20090234784Method of Providing Selected Content Items to a User - A method for providing selected content items to a user. The selection of content items is based on metadata pre-assigned to content items, typically authored content metadata, and on metadata generated and associated afterwards, called derived content metadata. Additionally, the selection of content items can be based also on context metadata, particularly derived context metadata. Derived metadata are automatically generated on the basis of derivation rules corresponding to algorithms to be applied to, e.g., the content of content items, authored content metadata and context metadata. User profiles can be used for improving the selection quality. A method is also disclosed for building and maintaining user profiles based on machine learning techniques.09-17-2009
20090234783VALUE FUNCTION REPRESENTATION METHOD OF REINFORCEMENT LEARNING AND APPARATUS USING THIS - Reinforcement learning is one of the intellectual operations applied to autonomously moving robots etc. It is a system having excellent sides, for example, enabling operation in unknown environments. However, it has the basic problem called the “incomplete perception problem”. A variety of solution has been proposed, but none has been decisive. The systems also become complex. A simple and effective method of solution has been desired.09-17-2009
20090234782METHOD AND APPARATUS FOR LOCATION EVALUATION AND SITE SELECTION - Method, apparatus and system for location evaluation and site selection, capable of effectively configuring the site network and evaluating the facility location by scientifically modeling and incorporating human knowledge are provided. In one aspect, geographic and demographic data associated with a plurality of locations and human knowledge comprising partial rating knowledge and pair-wise preference knowledge are used in a regression algorithm to construct a location evaluation model. The regression algorithm is further refined using active learning that identifies a plurality of pairs of locations to improve precision of the regression algorithm.09-17-2009
20090018982SEGMENTED MODELING OF LARGE DATA SETS - To provide efficient and effective modeling of data set, the data set is initially separated into several subsets which can then be processed independently. The subsets themselves are chosen to have some internal commonality, thus providing effective independent tools where possible. This commonality may include correlation between variables or interaction amongst the variables in the subset. Once separated, each subset is independently modeled, creating a subset model having predictive qualities related to the data subset. Next, the subset models themselves are aggregated to generate a overall final model. This final model is predictive of outcomes based upon all data in the data set, thus providing a more robust stable model.01-15-2009
20120233098Multiple Hypothesis Tracking - Embodiments described herein are directed to multiple hypothesis systems and methods for tracking observations that are domain agnostic and involves determining the probability that a given set of observations (i.e., a track) corresponds to a particular target, object or linked set of events. One embodiment described herein relates to cyber security tracking methods and systems.09-13-2012
20130166480NAVIGATION SYSTEM WITH POINT OF INTEREST CLASSIFICATION MECHANISM AND METHOD OF OPERATION THEREOF - A method of operation of a navigation system includes: generating a training data from a randomly sampled uncategorized point of interest; generating a trained classifier model by training a classifier model using the training data; generating a category identifier, a confidence score, or a combination thereof for an uncategorized point of interest using the trained classifier model; generating a categorized point of interest by assigning the category identifier to the uncategorized point of interest; calculating a weighted confidence score based on a weighted F-measure for the category identifier, a pair of the category identifier and the confidence score; and consolidating the categorized point of interest based on the weighted confidence score for the category identifier being meeting or exceeding a threshold for displaying on a device.06-27-2013
20130166482METHOD FOR DETERMINING A CORRECTION CHARACTERISTIC CURVE - A method for determining a correction characteristic curve for adapting a characteristic curve of an injection system, in which the correction characteristic curve includes at least one deviation of a measured characteristic curve from a setpoint characteristic curve, the at least one deviation including a sum tolerance of at least two components of the injection system, which have an effect on the characteristic curve.06-27-2013
20080301069SYSTEM AND METHOD FOR LEARNING BALANCED RELEVANCE FUNCTIONS FROM EXPERT AND USER JUDGMENTS - The present invention relates to systems and methods for determining a content item relevance function. The method comprises collecting user preference data at a search provider for storage in a user preference data store and collecting expert-judgment data at the search provider for storage in an expert sample data store. A modeling module trains a base model through the use of the expert-judgment data and tunes the base model through the use of the user preference data to learn a set of one or more tuned models. A measure (B measure) is designed to evaluate the balanced performance of tuned model over expert judgment and user preference. The modeling module generates or selects the content item relevance function from the tuned models with B measure as the selection criterion.12-04-2008
20120239599COMPUTER PRODUCT, DATA ANALYZING METHOD, AND DATA ANALYZING APPARATUS - A computer-readable medium stores a program that causes a computer, which has a memory device storing a set of measured values that include a set of positive case measured values for which an objective variable for an explanatory variable group of one or more explanatory variables represents a positive case and a set of negative case measured values for which the objective variable for the explanatory variable group represents a negative case, to execute a process. The process includes extracting randomly, a positive case measured value group and a negative case measured value group from the set of measured values such that the positive case measured values and the negative case measured values extracted are equivalent in number; and generating based on the positive case measured value group and the negative case measured value, a prediction equation that predicts the objective variable for a prediction algorithm.09-20-2012
20080294579LOW-POWER ANALOG-CIRCUIT ARCHITECTURE FOR DECODING NEURAL SIGNALS - A microchip for performing a neural decoding algorithm is provided. The microchip is implemented using ultra-low power electronics. Also, the microchip includes a tunable neural decodable filter implemented using a plurality of amplifiers, a plurality of parameter learning filters, a multiplier, a gain and time-constant biasing circuits; and analog memory. The microchip, in a training mode, learns to perform an optimized translation of a raw neural signal received from a population of cortical neurons into motor control parameters. The optimization being based on a modified gradient descent least square algorithm wherein update for a given parameter in a filter is proportional to an averaged product of an error in the final output that the filter affects and a filtered version of its input. The microchip, in an operational mode, issues commands to controlling a device using learned mappings.11-27-2008
20080294578DIAGNOSING INTERMITTENT FAULTS - A method and system for diagnosing any combination of persistent and intermittent faults. The behavior of a system under test is obtained by measuring or probing the system at a particular location(s). The predicted behavior of a modeled system corresponding to the system under test is investigated by drawing inferences based on at least conditional probabilities, prior observations and component models. The predictions are compared to their corresponding points in the system under test. A determination is made if a conflict exists between the measured behavior and the predicted behavior, and the conditional probabilities are adjusted to more and more accurately reflect the action fault(s) in the system under test. The conflicts or deviations between the obtained predicted behavior and the actual behavior are used to isolate the components of the system causing the faults.11-27-2008
20080294577Efficient Estimation of Events with Rare Occurrence Rates Using Taxonomies - Methods for predicting the click-through rates of Internet advertisements placed into web pages are disclosed. Specifically, a click-through rate prediction is generating using a hybrid system with two terms. The first term is constructed using a machine learning model that incorporates a limited number of important factors. The second term is constructed using a look-up table that is built using a complex statistical analysis of various web page and advertisement combinations. To construct the second term, the field of multi-level hierarchical modeling is used. Specifically, a tree-structured Markov model is used to process the training data and construct the adjustment factor look-up table. To reduce the complexity of the statistical analysis, Kalman-filters are used to estimate parameters in the traditional multi-level hierarchical models for scalability.11-27-2008
20090076988Method and system for optimal choice - A method and system for optimal choice is described. An inductive database system uses an integration of historical data and virtual data (in the form of intuitive rule-sets specified by an agent or plurality of agents) to make statistical recommendations for optimal choice. Filter mechanisms support the reporting of choice recommendations and user interaction with historical data. In the latter case, user interaction with a deductive interface allows for the testing of decision criteria or rule-sets against an historical database and empirical target results. The constant testing of ideas against an objective function provides an update methodology for a database of virtual data and provides a training methodology for the user. An example of picking stock investments is given.03-19-2009
20080288424Apparatus, Method, and Computer Program Product Providing Improved Identification of Suspect Entries in Transaction Data - The exemplary embodiments of the invention provide apparatus, systems, methods and computer program products for scoring entities in order to use the scoring for such tasks as identifying and prioritizing those entities that are candidates for further investigation, for example, from an audit or business control perspective. In an exemplary aspect of the invention, a method includes: providing transaction data having a plurality of pieces of information and an identification of a corresponding entity of a plurality of entities, wherein at least one piece of information of the plurality of pieces of information corresponds to each entity of the plurality of entities, wherein the transaction data comprises input data; computing at least one score for each entity of the plurality of entities by applying at least one statistical analysis technique to the input data, wherein the computed at least one score for a tested entity is indicative of at least one of a magnitude of deviation of the tested entity from a determined normal and repeated abnormal behavior of the tested entity; selecting zero or more entities of the plurality of entities by comparing at least one computed score of each entity with a specified threshold, wherein the selected zero or more entities comprise candidates for further investigation; and ordering the selected zero or more entities based on at least one computed score of each entity of the selected zero or more entities.11-20-2008
20100121792Directed Graph Embedding - Directed graph embedding is described. In one implementation, a system explores the link structure of a directed graph and embeds the vertices of the directed graph into a vector space while preserving affinities that are present among vertices of the directed graph. Such an embedded vector space facilitates general data analysis of the information in the directed graph. Optimal embedding can be achieved by measuring local affinities among vertices via transition probabilities between the vertices, based on a stationary distribution of Markov random walks through the directed graph. For classifying linked web pages represented by a directed graph, the system can train a support vector machine (SVM) classifier, which can operate in a user-selectable number of dimensions.05-13-2010
20090132446Support Vector Machines Processing System - An implementation of SVM functionality improves efficiency, time consumption, and data security, reduces the parameter tuning challenges presented to the inexperienced user, and reduces the computational costs of building SVM models. A computer program product for support vector machine processing in a computer system comprises computer program instructions for storing data, providing an interface to client software, building a support vector machine model on at least a portion of the stored data, based on a plurality of model-building parameters, estimating values for at least some of the model-building parameters, and applying the support vector machine model using the stored data to generate a data mining output.05-21-2009
20110184893ANNOTATING QUERIES OVER STRUCTURED DATA - A query may be received at a computing device and may include one or more terms. For each set of structured data tuples, a set of tokens may be determined from the terms of the query by the computing device based on attribute values of attributes associated with the structured data tuples in the set of structured data tuples. An annotated query may be determined from each of the sets of tokens. A probability score may be determined for each of the determined annotated queries. The annotated query having the highest determined probability score may be selected, and one or more structured data tuples may be identified from the structured data tuples that have attributes with attribute values that match one or more tokens of the selected annotated query.07-28-2011
20120179633IDENTIFICATION OF ATTRIBUTES AND VALUES USING MULTIPLE CLASSIFIERS - A body of text comprises a plurality of unknown attributes and a plurality of unknown values. A first classification sub-component labels a first portion of the plurality of unknown values as a first set of values, whereas a second classification sub-component labels a portion of the plurality of unknown attributes as a set of attributes and a second portion of the plurality of unknown values as a second set of values. Learning models implemented by the first and second classification subcomponents are updated based on the set of attributes and the first and second set of values. The first classification sub-component implements at least one supervised classification technique, whereas the second classification sub-component implements an unsupervised and/or semi-supervised classification technique. Active learning may be employed to provide at least one of a corrected attribute and/or corrected value that may be used to update the learning models.07-12-2012
20130024407TEXT CLASSIFIER SYSTEM - The present invention provides a method, and a system, for analysing a textual passage and classifying it against a number of predetermined categories. In the event that the text passage under analysis is not sufficiently similar to any of the predetermined categories then it will be classified as belonging to a further category.01-24-2013
20120191633System and Method For Failure Prediction For Artificial Lift Systems - A computer-implemented reservoir prediction system, method, and software are provided for failure prediction for artificial lift systems, such as sucker rod pump systems. The method includes a production well associated with an artificial lift system and data indicative of an operational status of the artificial lift system. One or more features are extracted from the artificial lift system data. Data mining is applied to the one or more features to determine whether the artificial lift system is predicted to fail within a given time period. An alert is output indicative of impending artificial lift system failures.07-26-2012
20110282815ASSOCIATION RULE MODULE FOR DATA MINING - A system, software module, and computer program product for performing association rule based data mining that improved performance in model building, good integration with the various databases throughout the enterprise, flexible specification and adjustment of the models being built, and flexible model arrangement and export capability. The software module for performing association rule based data mining in an electronic data processing system comprises: a model setup block operable to receive client input including information specifying a setup of a association rule data mining models, generate the model setup, generate parameters for the model setup based on the received information, a modeling algorithms block operable to select and initialize a association rule modeling algorithm based on the generated model setup, and a model building block operable to receive training data and build a association rule model using the training data and the selected association rule modeling algorithm.11-17-2011
20110282814METHODS AND SYSTEMS FOR IMPLEMENTING A COMPOSITIONAL RECOMMENDER FRAMEWORK - A compositional recommender framework using modular recommendation functions is described. Each modular recommendation function can use a discrete technology, such as using clustering, a database lookup, or other means. A first recommendation function can recommend to a user items, such as books to check out, automobiles to purchase, people to date, etc. Another modular recommendation function can be daisy chained with the first to recommend items that are similar or related to the first recommended items, such as users who have also checked out the same recommended book, trailers that can be towed by the recommended automobiles, or vacations booked by people that were recommended as people to date. The modular recommendation functions can be used to build customized recommendation engines for different industries.11-17-2011
20110282813SYSTEM AND METHOD FOR USING PATTERN RECOGNITION TO MONITOR AND MAINTAIN STATUS QUO - The present invention relates to a method of checking data gathered from a content source comprising: receiving initial data from the content source; training a data profiler to generate a set of trusted constraint modules, said training comprising (1) selecting constraint modules having parameters that are applicable to the initial data, (2) adjusting the parameters of the applicable constraint modules to conform with new data from the content source, (3) identifying non-stable constraint modules, and (4) generating a set of trusted constraint modules by removing the non-stable constraint modules; applying the set of trusted constraint modules to subsequently received data from the content source to determine whether the subsequently received data meets the parameters of the set of trusted constraint modules; and signaling a failure upon the subsequently received data failing to meet the parameters of the set of trusted constraint modules.11-17-2011
20110282812DYNAMIC PATTERN MATCHING OVER ORDERED AND DISORDERED DATA STREAMS - Architecture introduces a new pattern operator referred to as called an augmented transition network (ATN), which is a streaming adaptation of non-reentrant, fixed-state ATNs for dynamic patterns. Additional user-defined information is associated with automaton states and is accessible to transitions during execution. ATNs are created that directly model complex pattern continuous queries with arbitrary cycles in a transition graph. The architecture can express the desire to ignore some events during pattern detection, and can also detect the absence of data as part of a pattern. The architecture facilitates efficient support for negation, ignorable events, and state cleanup based on predicate punctuations.11-17-2011
20100268674Systems and Methods for Achieving PLMN Continuity When Moving Between Networks of Different Types Through Network Selection - Systems and methods for achieving PLMN continuity when moving between networks of different types through network selection are provided. When a mobile station moves from a first network type, such as cellular, to a second network type, such as GAN, if there is a PLMN discontinuity, this may result in a dropped call. In order to avoid this, networks for the first network type and the second network type are selected such that there is PLMN continuity. This can involve reselection of a different cellular network than one currently providing service to the mobile station.10-21-2010
20080281765Method of Ranking Politically Exposed Persons and Other Heightened Risk Persons and Entities - A method for ranking politically exposed persons and/or other persons and entities that pose a heightened risk based on their importance wherein an exposure index is determined for each person in the population as a function of the existence or absence of a relationship with each of the other members of the population and each of one or more exposure factors such as position held by the person, country in which the position is held, and source of information about the person. The politically exposed persons in the population are ranked in accordance with their respective exposure indexes. The population is sorted and a subset of the population containing those politically exposed persons having exposure indexes indicative or the highest likelihood of illicit financial activity is thereby identified.11-13-2008
20090106173LIMITED-MEMORY QUASI-NEWTON OPTIMIZATION ALGORITHM FOR L1-REGULARIZED OBJECTIVES - An algorithm that employs modified methods developed for optimizing differential functions but which can also handle the special non-differentiabilities that occur with the L04-23-2009
20080262984Field-Programmable Gate Array Based Accelerator System - Accelerator systems and methods are disclosed that utilize FPGA technology to achieve better parallelism and flexibility. The accelerator system may be used to implement a relevance-ranking algorithm, such as RankBoost, for a training process. The algorithm and related data structures may be organized to enable streaming data access and, thus, increase the training speed. The data may be compressed to enable the system and method to be operable with larger data sets. At least a portion of the approximated RankBoost algorithm may be implemented as a single instruction multiple data streams (SIMD) architecture with multiple processing engines (PEs) in the FPGA. Thus, large data sets can be loaded on memories associated with an FPGA to increase the speed of the relevance ranking algorithm.10-23-2008
20120290518INTEGRATED SEARCH AND ADAPTIVE DISCOVERY SYSTEM AND METHOD - An integrated search and adaptive discovery system and method integrates contents-based indexing and behavioral-based indexing of collections of computer-implemented objects to generate contextualized and/or personalized recommendations. The degree to which contextualization and/or personalization criteria are applied in generating recommendations can be tuned by the recommendation recipient. Personalization functions are applied that are informed by inferred interests and/or expertise, and personalization vectors can be transformed into a format executable by a search engine. Explanations may be provided to recommendation recipients as to why they received recommendations.11-15-2012
20120290516Habituation-compensated predictor of affective response - Creating a machine learning-based habituation-compensated predictor of a user's response to token instances representing stimuli that influence the user's affective state, comprising: receiving samples comprising temporal windows of token instances to which the user was exposed, wherein the token instances have overlapping instantiation periods; the samples further comprise data on previous instantiations of at least one of the token instances from the temporal windows; receiving target values corresponding to the temporal windows of token instances; the target values represent the user's responses to the token instances from the temporal windows of token instances; training the machine learning-based habituation-compensated predictor to predict the user's response to token instances, while accounting for the influence of the user's previous exposure to tokens; wherein the training uses the samples, the data on previous instantiations, and the corresponding target values11-15-2012
20120290517Predictor of affective response baseline values - Calculating a situation-dependent baseline value for a user response to token instances representing stimuli that influence the user's affective state, utilizing large time windows and rapid adjustments to changing situations, including: accessing a database storing annotations representing the user's response to token instances originating from multiple distinct token sources; calculating a first situation-dependent baseline value by weighting annotations retrieved from the database and associated with a first situation identifier, which are spread over a long period of time ‘T’; calculating a second situation-dependent baseline value by weighting annotations retrieved from the database and associated with a second situation identifier; wherein the difference between the first and second situation-dependent baseline values is significant, and the method rapidly adjusts to the situation change by exhibiting an extremely shorter transient time between the first and the second situation-dependent baselines than T/211-15-2012
20120290515Affective response predictor trained on partial data - Creating a machine learning-based affective response predictor of a user when there are significantly more samples than target values available for training, comprising: receiving samples comprising temporal windows of token instances to which the user was exposed; the token instances are spread over a long period of time; receiving intermittent target values corresponding to a subset of the temporal windows of token instances; the target values represent affective response annotations of the user; creating the machine learning-based affective response predictor of the user, by running a semi-supervised machine learning training procedure on the samples and the intermittent corresponding target values; wherein the machine learning-based affective response predictor is more accurate than a predictor created when training only on the samples that have corresponding target values, since it is capable of learning additional information from the samples comprising temporal windows of token instances without corresponding target values.11-15-2012
20120290514Methods for predicting affective response from stimuli - Creating a machine learning-based affective response predictor to predict a user's emotional state after being exposed to tokens representing stimuli that influence the user's affective state, comprising: receiving samples comprising temporal windows of token instances to which the user was exposed; the token instances are spread over a long period of time, and a subset of the token instances originate from same source and have overlapping instantiation periods; receiving target values, which represent affective response annotations of the user and correspond to the temporal windows of token instances; and creating the machine learning-based affective response predictor for the user, which compensates for non-linear effects resulting from the user being exposed to the subset of token instances originating from the same source and having overlapping instantiation periods, by running a machine learning training procedure on input data comprising the samples and the corresponding target values.11-15-2012
20120290513Habituation-compensated library of affective response - Generating a habituation-compensated library comprising a user's expected response to tokens representing stimuli that influence the user's affective state, the method comprising: receiving samples comprising temporal windows of token instances to which the user was exposed, wherein the token instances have overlapping instantiation periods; the samples further comprise data on previous instantiations of at least one of the token instances from the temporal windows; receiving target values corresponding to the temporal windows of token instances; the target values represent the user's response to the token instances from the temporal windows of token instances; training a machine learning-based user response model using the samples, the data on previous instantiations, and the corresponding target values; and analyzing the machine learning-based user response model to generate the habituation-compensated library, which accounts for the influence of the user's previous exposure to tokens11-15-2012
20120290512Methods for creating a situation dependent library of affective response - Generating a situation-dependent library comprising a user's expected response to tokens representing stimuli that influence the user's affective state, including: receiving samples comprising temporal windows of token instances to which the user was exposed, wherein the token instances have overlapping instantiation periods and are spread over a long period of time that spans different situations; wherein at least one token is expected to elicit from the user a noticeably different affective response in the different situations; receiving target values corresponding to the temporal windows of token instances; the target values represent the user's responses to the token instances from the temporal windows of token instances; training a machine learning-based user response model using the samples and the corresponding target values; and analyzing the machine learning-based user response model to generate the situation-dependent library comprising the user's expected response to tokens, which accounts for the variations in the user's affective response in the different situations.11-15-2012
20120290511Database of affective response and attention levels - A data structure stored in memory including: token instances representing stimuli that influence a user's affective state; the token instances are spread over a long period of time that spans different situations, and a plurality of the token instances have overlapping instantiation periods; data representing levels of user attention in some of the token instances used by an application program to improve the accuracy of a machine learning based affective response model for the user; annotations representing emotional states of the user; the annotations are spread over a long period of time that spans different situations; and linkage information between the token instances, the data representing levels of user attention, and the annotations.11-15-2012
20120290510MULTI-TASK MACHINE LEARNING USING FEATURES BAGGING AND LOCAL RELATEDNESS IN THE INSTANCE SPACE - A multi-task machine learning component learns a set of tasks comprising two or more different tasks based on a set of examples. The examples are represented by features of a set of features. The multi-task machine learning component comprises a digital processing device configured to learn an ensemble of base rules wherein each base rule is learned for a sub-set of the set of features and comprises a multi-task decision tree (MT-DT) having nodes comprising decision rules for tasks of the set of tasks. An inference component comprises a digital processing device configured to predict a result for at least one task of the set of tasks for an input item represented by features of the set of features using the learned ensemble of base rules.11-15-2012
20090018979MATH PROBLEM CHECKER - A problem checker architecture that monitors user progress during a problem-solving process and assists the user through the process (e.g., when requested) using common human methods of solving the problem. Assistance can be in the form of detecting errors during the process, and providing context-sensitive help information when the user gets stuck or makes a mistake. The problem checker can walk the user through the process of solving a math problem one step at a time allowing the user to learn to solve math problems according to a number of different methods. Rather than simply calculating and displaying the answer, the problem checker allows the user to attempt to solve math problems, providing direction only when asked and correction only when required. The problem checker can recognize multiple solution methods for many common math problems and guide the user to the solution via any of the methods.01-15-2009
20120005135RECOGNITION DICTIONARY TRAINING METHOD, SYSTEM, AND PROGRAM - The present invention provides a recognition dictionary training method, a system, and a program for allowing a computer to function as a recognition dictionary training system that does not cause a lowered recognition performance with regard to input vectors other than training data. In a recognition dictionary training system, an initial value setting means 01-05-2012
20120005134SPATIO-TEMPORAL LEARNING ALGORITHMS IN HIERARCHICAL TEMPORAL NETWORKS - A spatio-temporal learning node is a type of HTM node which learns both spatial and temporal groups of sensed input patterns over time. Spatio-temporal learning nodes comprise spatial poolers which are used to determine spatial groups in a set of sensed input patterns. The spatio-temporal learning nodes further comprise temporal poolers which are used to determine groups of sensed input patterns that temporally co-occur. A spatio-temporal learning network is a hierarchical network including a plurality of spatio-temporal learning nodes.01-05-2012
20110289027LEGACY SYSTEM SUPPORT - A system for adapting a legacy system to a new environment includes a method of learning the behavior of a legacy system and a method for replacing a legacy system.11-24-2011
20110289028PATTERN RECOGNITION DEVICE, PATTERN RECOGNITION METHOD, AND PATTERN RECOGNITION PROGRAM - A pattern recognition device executes feature selection using a feature selection table. High recognition performance is possible by dimensionally lowering an n-dimensional feature vector. A feature selecting table generating section for generating a feature selecting table in a manner so that when features with up to p11-24-2011
20110289026Matching Offers to Known Products - A method and apparatus for electronically matching an electronic offer to structured data for a product offering is disclosed. The structure data is reviewed and a dictionary of terms for each attribute from the structure data is created. Attributes in unstructured text may be determined. Each pair of the attributes (name and value) from the unstructured data and the structured data are obtained, the attribute pairs of the structured data and the unstructured data and compared and a similarity level is calculated for the matching the attribute pairs. The structured data pair that has the highest similarity score to the unstructured data pair is selected and returned.11-24-2011
20110289025LEARNING USER INTENT FROM RULE-BASED TRAINING DATA - The search intent co-learning technique described herein learns user search intents from rule-based training data and denoises and debiases this data. The technique generates several sets of biased and noisy training data using different rules. It trains each of a set of classifiers using different training data sets independently. The classifiers are then used to categorize the training data as well as any unlabeled data. The classified data confidently classified by one classifier is added to other training data sets, and the wrongly classified data is filtered out from the training data sets, so as to create an accurate training data set with which to train a classifier to learn a user's intent for submitting a search query string or targeting a user for on-line advertising based on user behavior.11-24-2011
20120016825DETECTOR CONFIGURATION APPARATUS, METHOD, AND PROGRAM - A detector configuration apparatus for configuring a detector that performs detection through a plurality of detection stages with different resolutions capable of objectively determining a detection target modality type to be detected in each stage. The detector is configured to detect to which of a plurality of attribute values an attribute of an object included in input data corresponds with respect to each of a plurality of modality types. A variation amount calculation unit obtains, based on a plurality of teacher data corresponding to each modality type used for training the detector, a representative value of variation between a plurality of teacher data with respect to each modality type, and a detection stage determination unit determines in which stage of the plurality of detection stages each modality type is to be detected based on the representative value of variation between the teacher data.01-19-2012
20110295782Clinical Decision Model - An embodiment of the invention provides a method for determining a patient-specific probability of disease. The method collects clinical parameters from a plurality of patients to create a training database. A fully unsupervised Bayesian Belief Network model is created using data from the training database; and, the fully unsupervised Bayesian Belief Network is validated. Clinical parameters are collected from an individual patient; and, such clinical parameters are input into the fully unsupervised Bayesian Belief Network model via a graphical user interface. The patient-specific probability of disease is output from the fully unsupervised Bayesian Belief Network model and sent to the graphical user interface for use by a clinician in pre-operative planning. The fully unsupervised Bayesian Belief Network model is updated using the clinical parameters from the individual patient and the patient-specific probability of disease.12-01-2011
20110295783Multiple Domain Anomaly Detection System and Method Using Fusion Rule and Visualization - The present invention discloses various embodiments of multiple domain anomaly detection systems and methods. In one embodiment of the invention, a multiple domain anomaly detection system uses a generic learning procedure per domain to create a “normal data profile” for each domain based on observation of data per domain, wherein the normal data profile for each domain can be used to determine and compute domain-specific anomaly data per domain. Then, domain-specific anomaly data per domain can be analyzed together in a cross-domain fusion data analysis using one or more fusion rules. The fusion rules may involve comparison of domain-specific anomaly data from multiple domains to derive a multiple-domain anomaly score meter for a particular cross-domain analysis task. The multiple domain anomaly detection system and its related method may also utilize domain-specific anomaly indicators of each domain to derive a cross-domain anomaly indicator using the fusion rules.12-01-2011
20110295776RESEARCH MISSION IDENTIFICATION - A system and method is described herein that automatically determines if a user of a search engine is conducting a research mission and then provides one or more research tools, one or more specialized searches, one or more directed ads, and/or one or more marketplace events responsive to determining that the research mission is being conducted. The automatic provision of various events and/or tools responsive to determination of the research mission can advantageously improve the experience of the user conducting the research mission.12-01-2011
20110295781Apparatus and Method for Improved Classifier Training - A non-transitory computer readable storage medium includes instructions to maintain an original training set of labeled documents, where the labeled documents correspond to a variety of topics. A new labeled document corresponding to a new topic is received. The original training set of labeled documents is modulated such that the new labeled document is over-represented with respect to the original training set. This results in a modulated training set. A classifier is trained with the modulated training set to form a trained classifier.12-01-2011
20110295780Apparatus and Method for Personalized Delivery of Content from Multiple Data Sources - A non-transitory computer readable storage medium includes instructions to collect explicit feedback from a user regarding user content preferences. Multiple data sources are monitored. Topics associated with the multiple data sources are classified. The importance of the topics to the user is characterized. Content is delivered to the user when a selected topic exceeds an importance threshold for the user. Implicit feedback from the user that characterizes refined user content preferences is tracked. The instructions to characterize the importance of topics evaluates the explicit feedback and the implicit feedback.12-01-2011
20110295779REGULAR EXPRESSION MATCHING METHOD AND SYSTEM - The present invention discloses a regex matching method and system, and relates to the field of computer technologies. The method includes: sorting multiple regexes into several regex groups, where all regexes in one regex group include a common string, which is known as a generic string; compiling each regex group into a DFA, and setting up a correlation between the generic string of each regex group and the DFA; matching to-be-matched data streams with the generic string respectively, and using the matched generic string as a matched string; obtaining a DFA corresponding to the matched string; and performing regex matching for the to-be-matched data streams according to the DFA, and outputting a matching result. The embodiments of the present invention shorten the data loading process, decrease the time consumed by data loading, and improve the matching performance.12-01-2011
20110295778INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND PROGRAM - There is provided an information processing apparatus including a data pool generation section which generates an unknown data pool, a learning sample collection section which randomly collects a plurality of learning samples from the unknown data pool, a classifier generation section which generates a plurality of classifiers using the learning samples, an output feature quantity acquisition section which associates with the data, for each piece of the data, a plurality of output values, which are obtained by inputting the data into the plurality of classifiers to identify the data, as an output feature quantity represented in an output feature quantity space different from the feature quantity space, and a classification section which classifies each piece of the data into any one of a predetermined number of the classes based on the output feature quantity.12-01-2011
20110295777METHOD FOR BUILDING ADAPTIVE SOFT SENSOR - The invention discloses a method for building adaptive soft sensor. The method comprises the following steps. The input and schedule vectors are constructed, and a novel learning algorithm that uses online subtractive clustering is used to recursively update the structure and parameters of a local model network. Three rules are proposed for updating centers and local model coefficients of existing clusters, for generating new clusters and new models as well as for merging existing clusters and their corresponding models. Once verified, the online inferential model can be created to generate the predicted value of process. Thus, it does not need much memory space to process the method and can be easily applied to any other machine.12-01-2011
20110295775ASSOCIATING MEDIA WITH METADATA OF NEAR-DUPLICATES - Techniques for identifying near-duplicates of a media object and associating metadata of the near-duplicates with the media object are described herein. One or more devices implementing the techniques are configured to identify the near duplicates based at least on similarity attributes included in the media object. Metadata is then extracted from the near-duplicates and is associated with the media object as descriptors of the media object to enable discovery of the media object based on the descriptors.12-01-2011
20110295774Training SVMs with Parallelized Stochastic Gradient Descent - Techniques for training a non-linear support vector machine utilizing a stochastic gradient descent algorithm are provided. The computations of the stochastic gradient descent algorithm are parallelized via a number of processors. Calculations of the stochastic gradient descent algorithm on a particular processor may be combined according to a packing strategy before communicating the results of the calculations with the other processors.12-01-2011
20110173141METHOD AND APPARATUS FOR HYBRID TAGGING AND BROWSING ANNOTATION FOR MULTIMEDIA CONTENT - A computer program product and embodiments of systems are provided for annotating multimedia documents. The computer program product and embodiments of the systems provide for performing manual and automatic annotation.07-14-2011
20110302116DATA PROCESSING DEVICE, DATA PROCESSING METHOD, AND PROGRAM - A data processing device including a learning section which expresses user movement history data obtained as learning data as a probability model which expresses activities of a user and learns parameters of the model; a destination and stopover estimation section which estimates a destination node and a stopover node from state nodes of the probability model; a current location estimation section which inputs the user movement history data in the probability model and estimates a current location node which is equivalent to the current location of the user; a searching section which searches for a route from the current location of the user to a destination using information on the estimated destination node and stopover node and the current location node and the probability model obtained by learning; and a calculating section which calculates an arrival probability and a necessary time to the searched destination.12-08-2011
20110302117INTERESTINGNESS RECOMMENDATIONS IN A COMPUTING ADVICE FACILITY - The present disclosure provides a recommendation to a user through a computer-based advice facility, comprising collecting topical information, wherein the collected topical information includes an interestingness aspect; filtering the collected topical information based on the interestingness aspect; determining an interestingness rating from the collected topical information, wherein the determining is through the computer-based advice facility; and providing a user with the recommendation related to the topical information based on the interestingness rating.12-08-2011
20110302115METHOD AND DEVICE FOR INFORMATION RETRIEVAL - A method of information retrieval, comprising: determining, using a microprocessor, Q generative models (λ) in accordance with Probabilistic Latent Semantic Indexing (PLSI), said Q generative models being determined in an offline training; receiving a user query (q); choosing, using said microprocessor, N generative models out of the Q generative models, with N≦Q; determining, using said microprocessor, a content item (d) based on said query and a combination of the N generative models.12-08-2011
20110302114Systems and Methods For Turbo On-Line One-Class Learning - Methods for one-class learning using support vector machines from a plurality of data batches are provided. A first support vector machine is learned from the plurality of data batches by a processor. A new data batch is received by the processor and is classified by the first support vector machine. If a non-zero loss classification occurs a new support vector machine is trained using the first support vector machine and the new data batch only. Data batches can be discarded if they are represented by the current support vector machine or after being used for training an updated support vector machine. Weighing factors applied to update the first support vector machine depend upon a parameter which is optimized iteratively. Support vectors do not need to be recalculated. A classifier is learned in a number of stages equal to the number of data batches processed on-line.12-08-2011
20110302113MONITORING RELATIONSHIPS BETWEEN DIGITAL ITEMS ON A COMPUTING APPARATUS - Systems and methods are described herein that facilitate file management on a computing device and/or across multiple computing devices. Actions of a user with respect to digital items on a computing device can be monitored and utilized to build a relationship table, wherein the relationship table comprises identities of digital items and data that describes relationships between particular digital items. Such a relationship table is analyzed to provide a user with information pertaining to relationships between digital items captured in the relationship table. Relationship tables from different devices can be merged to analyze relationships between digital items across devices.12-08-2011
20110302112FORECASTING THROUGH TIME DOMAIN ANALYSIS - Embodiments include methods, apparatus, and systems for forecasting using a time domain analysis. One embodiment is a computer implemented method that receives plural cycle lengths identified in time series data and builds a model using a time domain analysis of the time series data. The model is used to predict future events or future data points.12-08-2011
20110302111MULTI-LABEL CLASSIFICATION USING A LEARNED COMBINATION OF BASE CLASSIFIERS - Multi-label classification is performed by (i) applying a set of trained base classifiers to an object to generate base classifier label prediction sets comprising subsets of a set of labels; (ii) constructing a set of second level features including at least one second level feature defined by a predetermined combination of two or more of the base classifier label prediction sets; and (iii) applying a second level classifier to label the object with a set of one or more labels comprising a subset of the set of labels, labeling being based on the set of second level features. The multi-label classifier is trained by: (iv) applying operations (i) and (ii) to labeled training objects of a set of labeled training objects to generate training metadata comprising sets of second level features for the labeled training objects; and (v) training the second level classifier using the training metadata.12-08-2011
20090171872Selection of head and neck cancer patients for treatment with drugs targeting EGFR pathway - Methods using mass spectral data analysis and a classification algorithm provide an ability to determine whether a head and neck squamous cell carcinoma (HNSCC) patient is likely to benefit from a drug targeting an epidermal growth factor receptor pathway, including small molecule epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) and monoclonal antibody EGFR inhibitors.07-02-2009
20100145893GENOMIC CLASSIFICATION OF NON-SMALL CELL LUNG CARCINOMA BASED ON PATTERNS OF GENE COPY NUMBER ALTERATIONS - The invention is directed to methods and kits that allow for classification of non-small cell lung carcinoma tumors and cell lines according to genomic profiles, and methods of diagnosing, predicting clinical outcomes, and stratifying patient populations for clinical testing and treatment using the same.06-10-2010
20110191277AUTOMATIC DATA MINING PROCESS CONTROL - A data mining system includes a planning and learning module which receives as input a knowledge model and a set of goals and automatically produces as output a plurality of plans. The system includes a data mining processing unit which receives the plans as instructions and automatically creates results which are provided back to the planning and learning module as feedback. A method for data mining includes the steps of receiving as input at a planning and learning module a knowledge model and a set of goals. There is the step of automatically producing as output of the planning and learning module a plurality of plans from the input. There is the step of receiving by a data mining processing unit the plans as instructions. There is the step of automatically creating results by the data mining processing unit. There is the step of providing back to the planning and learning module the results as feedback.08-04-2011
20090030857MULTIATTRIBUTE SPECIFICATION OF PREFERENCES ABOUT PEOPLE, PRIORITIES, AND PRIVACY FOR GUIDING MESSAGING AND COMMUNICATIONS - The present invention relates to a system and methodology to facilitate multiattribute adjustments and control associated with messages and other communications and informational items that are directed to a user via automated systems. An interface, specification language, and controls are provided for defining a plurality of variously configured groups that may attempt to communicate respective items. Controls include the specification of priorities and preferences as well as the modification of priorities and preferences that have been learned from training sets via machine learning methods. The system provides both a means for assessing parameters used in the control of messaging and communications and for the inspection and modification of parameters that have been learned autonomously.01-29-2009
20110145176GENE EXPRESSION PROFILES TO PREDICT BREAST CANCER OUTCOMES - Methods for classifying and for evaluating the prognosis of a subject having breast cancer are provided. The methods include prediction of breast cancer subtype using a supervised algorithm trained to stratify subjects on the basis of breast cancer intrinsic subtype. The prediction model is based on the gene expression profile of the intrinsic genes listed in Table 1. This prediction model can be used to accurately predict the intrinsic subtype of a subject diagnosed with or suspected of having breast cancer. Further provided are compositions and methods for predicting outcome or response to therapy of a subject diagnosed with or suspected of having breast cancer. These methods are useful for guiding or determining treatment options for a subject afflicted with breast cancer. Methods of the invention further include means for evaluating gene expression profiles, including microarrays and quantitative polymerase chain reaction assays, as well as kits comprising reagents for practicing the methods of the invention.06-16-2011
20100036781APPARATUS AND METHOD PROVIDING RETRIEVAL OF ILLEGAL MOTION PICTURE DATA - Provided are an apparatus and method for detecting illegal motion picture data. The apparatus includes a key frame extractor for extracting a plurality of key frames from motion picture data, a characteristic value file generator for detecting characteristic values of the extracted key frames and generating a characteristic value file, and an illegality determiner for measuring degree of similarity between a previously stored learning model file and the characteristic value file and determining whether or not the motion picture data is legal according to the degree of similarity.02-11-2010
20110264613METHODS, APPARATUS AND SYSTEMS USING PROBABILISTIC TECHNIQUES IN TRENDING AND PROFILING - An embodiment of the present invention provides a mobile device, comprising a processor adapted to use probabilistic techniques in trending and profiling of a user of the mobile device's behavior in order to offer recommendations by detecting patterns in the user behavior over time and thereby enabling said mobile device to predict what the user is likely to do on a given day or what the user intends to accomplish in an action that has begun.10-27-2011
20110218952SOUND IDENTIFICATION SYSTEMS - We describe a digital sound identification system, the system comprising: non-volatile memory for storing a Markov model; stored program memory storing processor control code; a sound data input; a processor coupled to said sound data input, to said working memory, and to said stored program memory for executing said processor control code, and wherein said processor control code comprises code to: input, from said sound data input, first sample sound data for a first sound to be identified, said first sample sound data defining first sample frequency domain data, said first sample frequency domain data defining an energy of said first sample in a plurality of frequency ranges; generate a first set of mean and variance values for at least a first Markov model of said first sample sound from said first sample frequency domain data; store said first Markov model in said non-volatile memory; input interference sound data defining interference frequency domain data; adjust said mean and variance values of said first Markov model using said interference frequency domain data; input third sound data defining third sound frequency domain data; determine a probability of said third sound frequency domain data fitting at least said first Markov model; and output sound identification data dependent on said probability.09-08-2011
20110218945TRAINING WITH COMPLEX EVENT PROCESSING ENGINE TO IDENTIFY SEMANTIC MEANING OF VIRTUAL WORLD OBJECT STATE CHANGES - Techniques for training a system to identify state changes in objects in virtual worlds. Base events transmitted by a virtual world engine are observed. Statistical analysis of the observed base events is performed. Based at least in part on this statistical analysis, a computer processor determines that a group of one or more of the observed base events is correlated to a first identified higher-level event. Optionally, the determination is based in part on a frequency of occurrence of the group of base events, on generated rules, or both. A candidate higher-level event including the group of base events thus determined is stored. User input is received about the candidate higher-level event. If so specified by the received user input, the candidate higher-level event is stored as a second identified higher-level event. As a result, the system is advantageously trained to identify higher-level events which represent abstract situations.09-08-2011
20110218950METHOD, SYSTEM, AND COMPUTER-ACCESSIBLE MEDIUM FOR CLASSIFICATION OF AT LEAST ONE ICTAL STATE - An exemplary methodology, procedure, system, method and computer-accessible medium can be provided for receiving physiological data for the subject, extracting one or more patterns of features from the physiological data, and classifying the at least one state of the subject using a spatial structure and a temporal structure of the one or more patterns of features, wherein at least one of the at least one state is an ictal state.09-08-2011
20110218951SYSTEM AND METHOD FOR PUSHING DATA TO A MOBILE DEVICE - A method for handling information requests from mobile devices includes a memory, a state prediction module, and a push module. The memory is operable to store data requests received from the mobile devices. The state prediction module is operable to access the memory to predict forecasted data requests for a mobile device based on the stored data requests. The push module is operable to receive the forecasted data requests from the state prediction module and in response request and receive response data related to the forecasted data requests and prepare the response data for transmission to the mobile device over a wireless network.09-08-2011
20110218949METHOD OF OPTIMIZING DATA TRAINING IN SYSTEM INCLUDING MEMORY DEVICES - In one embodiment, a method of performing data training in a system including a memory controller and at least a first memory device including a group of memory banks is disclosed. The method includes providing a plurality of enabling states for the group of memory banks, wherein each enabling state is different and for each enabling state a set of the memory banks of the group is enabled and any remaining of the memory banks of the group are not enabled. The method further includes performing a first data training procedure that includes a series of first data training operations for the first memory device, each data training operation being performed for a different one of the plurality of enabling states, generating a noise profile based on the series of first data training operations, statistically analyzing the noise profile to select a reference enabling state of the group of memory banks, and performing a second data training procedure for the first memory device using the reference enabling state. As a result, the operating speed and reliability of the system including the memory device may be improved.09-08-2011
20110218948METHODS FOR DETECTING SPAMMERS AND CONTENT PROMOTERS IN ONLINE VIDEO SOCIAL NETWORKS - The present invention relates to a method for detecting video spammers and promoters in online video social systems. Using attributes based on the user's profile, the user's social behavior in the system, and the videos posted by the user as well as the target (responded) videos, the feasibility of applying a supervised learning method to identify polluters (spammers and promoters) is investigated.09-08-2011
20110218947ONTOLOGICAL CATEGORIZATION OF QUESTION CONCEPTS FROM DOCUMENT SUMMARIES - Electronic documents are analyzed to identify assertions, which are inverted to generate questions that may be answered by the assertions. A document or a corpus of electronic documents may be analyzed to identify entities and relationships among entities within the text of the document(s). Assertions are identified based on the entities and relationships among the entities. Each assertion represents a fact about an entity, and a group of assertions represents a summary of the document or document corpus. The assertions are inverted to generate questions that may be answered by the assertions. The questions may be further analyzed to identify relevant concepts and topics and to cluster the questions around the concepts and topics. A combined graph may also be generated that facilitates traversal among topics, concepts, questions, assertions, document summaries, and documents.09-08-2011
20110218946PRESENTING CONTENT ITEMS USING TOPICAL RELEVANCE AND TRENDING POPULARITY - A user may request a presentation of a content item set, such as a social network comprising a set of status messages or an image database comprising a set of images. However, the volume and diversity of content items of the content item set may reduce the interest of the user in the presented content items. The potential interest of the user in the presented content items may be improved by selecting content items that are associated with one or more topics of potential interest to the user, and having a positive trending popularity among users of the content item set. Moreover, the interaction of the user with a presented content item may be monitored and used to determine the interest of the user in the topics associated with the presented content item and the popularity of the content item.09-08-2011
20090210365System and method for combining hetergeneous predictors with an application to survival anaylsis - A method, a system, and a computer-readable medium for predicting a risk in a survival analysis for a plurality of individuals characterized by at least one predictor are disclosed. A method for estimating risk order of an individual, given information about a set of individuals, characterized by one or many predictors, and provided that direction of association between each predictor and the risk order is known, comprising the step of comparing the individual with each individual within the set of individuals, and estimating risk of individual based on set comparisons.08-20-2009
20110191274Deep-Structured Conditional Random Fields for Sequential Labeling and Classification - Described is a technology by which a deep-structured (multiple layered) conditional random field model is trained and used for classification of sequential data. Sequential data is processed at each layer, from the lowest layer to a final (highest) layer, to output data in the form of conditional probabilities of classes given the sequential input data. Each higher layer inputs the conditional probability data and the sequential data jointly to output further probability data, and so forth, until the final layer which outputs the classification data. Also described is layer-by-layer training, supervised or unsupervised. Unsupervised training may process raw features to minimize average frame-level conditional entropy while maximizing state occupation entropy, or to minimize reconstruction error. Also described is a technique for back-propagation of error information of the final layer to iteratively fine tune the parameters of the lower layers, and joint training, including joint training via subgroups of layers.08-04-2011
20110191273Evaluating ontologies - A method for providing an evaluation/verification of the correctness of an ontology is described. The method includes loading a first ontology associated with a first rule set. an extended ontology and an extended rule set are generated based at least in part on the first ontology and the first rule set. The extended rule set is applied to the extended ontology. The method also includes determining (e.g., by a data processor) a correctness of the extended ontology. Results are generated which include the correctness. Apparatus and computer readable media are also described.08-04-2011
20090177597SYSTEMS, METHODS AND COMPUTER PRODUCTS FOR PROFILE BASED IDENTITY VERIFICATION OVER THE INTERNET - Systems, methods and computer products for profile-based identity verification over the Internet. Exemplary embodiments include a system including an activity classifier configured to receive Internet activity input including email, chat, browser and voice over Internet Protocol (VoIP) logs/streams, an email profiler, a chat, a browser profiler, a voice over Internet Protocol (VoIP) logs/streams profiler, wherein the profilers are configured to extract values from the Internet Activity input attributes from the data set, a score calculator configured to receive the attributes and calculate the score of the data set, a categorization engine configured to receive the score from the score calculator and map the data set to an individual or class of individuals based on the value of the score and on a database of activity-specific attributes and an application configured to place weights on the activity specific and generic attributes to define a score function from the score.07-09-2009
20100100512METHOD AND ARRANGEMENT FOR RANKING OF LIVE WEB APPLICATIONS - A method of ranking a plurality of live web applications of a communication device is disclosed. The method comprises receiving at least one data stream, each having a content and associated with a corresponding one of the plurality of live web applications, and evaluating the content of the at least one data stream using machine-learning algorithms. The method further comprises updating each of the corresponding live web applications based on the at least one data stream and determining for each of the corresponding live web applications whether any user reaction occurs with the corresponding live web application in association with the updating of the corresponding live web application. The method comprises ranking the plurality of live web applications relative to each other based at least on the evaluation of the content of the at least one data stream and the determinations of whether any user reaction occurred. Corresponding computer program product, arrangement and communication device are also disclosed.04-22-2010
20090319450PROTEIN SEARCH METHOD AND DEVICE - A protein search method for searching for, as a target protein, a protein having direct or indirect relevance to information based on protein representation profiling data acquired by means of proteome analysis includes: determining, as a target protein, a protein that is relevant to the information based on significance of proteins obtained by using supervised learning from the information and the protein representation in the profiling data; and evaluating the performance of the target protein by means of evaluation data.12-24-2009
20090125463TECHNIQUE FOR CLASSIFYING DATA - Provided is a system that generates models for classifying input data into a plurality of classes on the basis of training data previously classified into the plurality of classes. The system includes a sampling unit and a learning unit. The sampling unit samples, from the training data, a plurality of datasets each including a predetermined number of elements classified into a minority class and a corresponding number of elements classified into a majority class, the corresponding number being determined in accordance with the predetermined number. The learning unit learns each of a plurality of models for classifying the input data into the plurality of classes, by using a machine learning technique on the basis of each of the plurality of sampled datasets.05-14-2009
20100114802SYSTEM AND METHOD FOR AUTOMATICALLY DISTINGUISHING BETWEEN CUSTOMERS AND IN-STORE EMPLOYEES - An approach that automatically distinguishes between in-store customers and in-store employees is provided. In one embodiment, there is a learning tool configured to construct a model for an in-store employee; and a classifying tool, further comprising matching tool configured to: match attributes between a particular person and the constructed models for an in-store employee, the classifying tool configured to: classify persons into categories of employees and customers based on amount of matching attributes between a particular person and the model for an in-store employee.05-06-2010
20090150310PORTABLE COMMUNICATION DEVICE, IN-VEHICLE COMMUNICATION DEVICE, AND COMMUNICATION SYSTEM - A communication system has: a portable communication device; and an in-vehicle communication device that communicates with the portable communication device through synchronous communication. The potable communication device transmits dummy signals simulating the vehicle information to the in-vehicle communication device in a first period. The in-vehicle communication device learns and records a first filter coefficient for filtering the dummy signals that are received in the first period and a second filter coefficient for filtering signals that are received from the sensors in a second period in which the portable communication device do not transmit the dummy signals, and the in-vehicle communication device calculates a third filter coefficient for removing noises on signals received from the sensors based on the first and second filter coefficients and then filters the received signals using the third filter coefficient.06-11-2009
20090150312Systems And Methods For Analyzing Disparate Treatment In Financial Transactions - Systems and methods are provided for analyzing disparate treatment in financial transactions. Data processing software instructions may be used to process lending-related data to identify a plurality of primary factors and one or more secondary factors for use making a lending-related decision. Model facilitation software instructions may be used to receive one or more relationships between the primary factors and the one or more secondary factors, wherein the relationships define criteria in which one or more positive secondary factors will compensate for a negative primary factor in making the lending-related decision. Model generation software instructions may be used to analyze lending-related data based on the primary factors, secondary factors and the one or more relationships.06-11-2009
20100268673ASSOCIATE MEMORY LEARNING AGENT TECHNOLOGY FOR TRAVEL OPTIMIZATION AND MONITORING - A method for assisting with evaluating travel related information. The method may involve defining a plurality of entity types for categorizing different types of travel related information. A data mining tool may be used to search at least one database for stored travel related information, and to denote specific items of the travel related information as entities. The data mining tool may be used to populate an associative learning memory with the entities and to store the entities in the associative learning memory. An entity analytics engine may be used to assist a user in searching the associative learning memory for specific ones of the entities that are at least one of identical or similar to specific travel related information provided by the user. Retrieved ones of the entities may be displayed for evaluation by the user.10-21-2010
20090094175INTRUSIVE SOFTWARE MANAGEMENT - Intrusion features of a landing page associated with sponsored content are identified. A feature score for the landing page based on the identified intrusion features is generated, and if the feature score for the landing page exceeds a feature threshold, the landing page is classified as a candidate landing page. A sponsor account associated with the candidate landing page can be suspended, or sponsored content associated with the candidate landing page can be suspended.04-09-2009
20090150308MAXIMUM ENTROPY MODEL PARAMETERIZATION - Described is a technology by which a maximum entropy model used for classification is trained with a significantly lesser amount of training data than is normally used in training other maximum entropy models, yet provides similar accuracy to the others. The maximum entropy model is initially parameterized with parameter values determined from weights obtained by training a vector space model or an n-gram model. The weights may be scaled into the initial parameter values by determining a scaling factor. Gaussian mean values may also be determined, and used for regularization in training the maximum entropy model. Scaling may also be applied to the Gaussian mean values. After initial parameterization, training comprises using training data to iteratively adjust the initial parameters into adjusted parameters until convergence is determined.06-11-2009
20090150311Action based learning - A set of sequences of sensed input patterns associated with a set of actions is generated by performing at least a first action on data derived from a real-world system. A subset of the sequences of sensed input patterns that form a group associated with the first action is determined. A new sequence of sensed input patterns is received. A first value which indicates the probability that the new sequence of sensed input patterns is associated with the first action based on the subset of sequences of sensed input patterns is determined and stored in a memory associated with the computer system.06-11-2009
20100125540System And Method For Providing Robust Topic Identification In Social Indexes - A computer-implemented method for providing robust topic identification in social indexes is described. Electronically-stored articles and one or more indexes are maintained. Each index includes topics that each relate to one or more of the articles. A random sampling and a selective sampling of the articles are both selected. For each topic, characteristic words included in the articles in each of the random sampling and the selective sampling are identified. Frequencies of occurrence of the characteristic words in each of the random sampling and the selective sampling are determined. A ratio of the frequencies of occurrence for the characteristic words included in the random sampling and the selective sampling is identified. Finally, for each topic, a coarse-grained topic model is built, which includes the characteristic words included in the articles relating to the topic and scores assigned to those characteristic words.05-20-2010
20110264612Automatic Rule Discovery From Large-Scale Datasets to Detect Payment Card Fraud Using Classifiers - A set of payment card transactions including a sparse set of fraudulent transactions is normalized, such that continuously valued literals in each of the set of transactions are transformed to discrete literals. The normalized transactions are used to train a classifier, such as a neural network, such that the classifier is trained to classify transactions as fraudulent or genuine. The fraudulent transactions in the set of payment card transactions are clustered to form a set of prototype transactions. Each of the discrete literals in each of the prototype transactions is expanded using sensitivity analysis using the trained classifier as an oracle, and a rule for identifying fraudulent transactions is generated for each prototype transaction based on the transaction's respective expanded literals.10-27-2011
20110264611PRESENTING AN INTERACTIVE GUIDANCE STRUCTURE IN A COLLABORATIVE ENVIRONMENT - A collaborative work environment is provided that supports collaboration among users for performance of a people service that is associated with ad-hoc activities. An information base is provided that includes information relating to responsibilities of the users and work items for the ad-hoc activities. An interactive guidance structure is presented in the collaborative environment to guide actions of the users with respect to the work items. Materials produced as a result of the actions to update the information base are collected.10-27-2011
20110264610Address Data Learning and Registration Within a Distributed Virtual Bridge - Systems and methods to forward data frames are provided. A particular apparatus may include a plurality of server computers and a distributed virtual bridge. The distributed virtual bridge may include a plurality of bridge elements coupled to the plurality of server computers and configured to forward a data frame between the plurality of server computers. The plurality of bridge elements may further be configured to automatically learn address data associated with the data frame. A controlling bridge may be coupled to the plurality of bridge elements. The controlling bridge may include a global forwarding table that is automatically updated to include the address data and is accessible to the plurality of bridge elements.10-27-2011
20090119234INTERACTIVE MACHINE LEARNING ADVICE FACILITY - In embodiments of the present invention improved capabilities are described for helping a user make a decision through the use of a machine learning facility. The process may begin with an initial question being received by the machine learning facility from the user. The user may then be provided with a dialog consisting of questions from the machine learning facility and answers provided by the user. The machine learning facility may then provide a decision to the user based on the dialog and pertaining to the initial question, such as a recommendation, a diagnosis, a conclusion, advice, and the like. In embodiments, future questions and decisions provided by the machine learning facility may be improved through feedback provided by the user.05-07-2009
20100235308TEXT ANALYSIS DEVICE AND METHOD AND PROGRAM - A text analysis device includes a storage unit configured to store opinions of users who participate in a discussion about a predetermined theme as text data and author information for specifying authors of the text data, a feature quantity data generation unit configured to generate feature quantity data of the text data stored in the storage unit, an observation time-series signal generation unit configured to generate observation time-series signals based on information obtained by performing a predetermined process with respect to the feature quantity data, a change point detection unit configured to detect a change point of the discussion based on the observation time-series signals, and an influence specifying unit configured to specify an opinion having influence on an opinion corresponding to the specified text data out of opinions of the discussion based on the detected change point and the author information.09-16-2010
20090182691METHOD AND SYSTEM OF ENHANCING GANGLION CELL FUNCTION TO IMPROVE PHYSICAL PERFORMANCE - A method and a system of enhancing ganglion cell function using a gaming environment corresponding to a physical activity. The method and system may be used to implement one or more processes to improve a person's visual processing profile. In particular, the method and system may be used to improve a player's skill in the corresponding physical activity.07-16-2009
20120109862USER DEVICE AND METHOD OF RECOGNIZING USER CONTEXT - A method and user device for recognizing a user context are provided. The method includes: recognizing at least one behavior generated from an object by analyzing a signal obtained by at least one sensor from among a plurality of sensors included in a user device; and recognizing a current context of the user by analyzing a pattern of the at least one behavior. According to the method, a behavior of a user of a user device such a smart phone may be analyzed in real time and an appropriate service for the behavior may be provided according to the result of the analysis.05-03-2012
20120109861INFORMATION PROCESSING APPARATUS, PROCESSING METHOD THEREFOR, AND NON-TRANSITORY COMPUTER-READABLE STORAGE MEDIUM - An information processing apparatus creates, for each of a plurality of nodes, a query to be executed for a learning pattern input to the node; inputs a plurality of learning patterns to a root node of the plurality of nodes; executes, for the learning pattern input to each node, the query created for the node; determines whether the query has been effectively executed for the individual learning pattern input to each node; distributes and inputs, to a lower node of each node, an individual learning pattern for which it has been determined in the determining that the query was effectively executed in the node; deletes a learning pattern for which it has been determined in the determining that the query was not effectively executed in each node; and stores an attribute of the learning pattern input to a terminal node of the plurality of nodes in association with the node.05-03-2012
20120109860Enhanced Training Data for Learning-To-Rank - Training data is used by learning-to-rank algorithms for formulating ranking algorithms. The training data can be initially provided by human judges, and then modeled in light of user click-through data to detect probable ranking errors. The probable ranking errors are provided to the original human judges, who can refine the training data in light of this information.05-03-2012
20120109859Scalable Ontology Extraction - Techniques for facilitating learning of one or more ontological rules of a resource description framework database are provided. The techniques include obtaining ontology vocabulary from a resource description framework database, generating a rule hypothesis by incrementally building upon a previously learnt rule from the database by adding one or more predicates to the previously learnt rule, performing a constraint check on the generated rule hypothesis by determining compatibility with each previously learnt rule to ensure that a complete rule set including each previously learnt rule and the generated rule hypothesis is consistent, validating the rule hypothesis as a rule using one or more association rule mining techniques to determine validity of the rule hypothesis against the database, and applying the rule to the database to infer one or more facts from the database to facilitate learning of one or more additional ontological rules.05-03-2012
20120109858Search with Joint Image-Audio Queries - Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for processing joint image-audio queries. In one aspect, a method includes receiving, from a client device, a joint image-audio query including query image data and query audio data. Query image feature data is determined from the query image data. Query audio feature data is determined from the audio data. The query image feature data and the query audio feature data are provided to a joint image-audio relevance model trained to generate relevance scores for a plurality of resources, each resource including resource image data defining a resource image for the resource and text data defining resource text for the resource. Each relevance score is a measure of the relevance of corresponding resource to the joint image-audio query. Data defining search results indicating the order of the resources is provided to the client device.05-03-2012
20130218813CLASSIFICATION RELIABILITY PREDICTION - A method, apparatus and product useful for classification reliability prediction. The method being a computer-implemented method performed by a processor, the method comprising: obtaining a prediction of a label for a dataset made by a classifier tool, wherein the classifier tool is aimed at predicting the label based on a classification model and in view of a set of features defining the dataset; obtaining a reliability prediction of a reliability label relating to the prediction of the classifier tool based on a reliability classifier tool, wherein the reliability classifier tool is aimed at predicting the reliability label based on a classification model and in view of a second set of features; and outputting to a user the label prediction and an associated reliability prediction.08-22-2013
20120036093Waveform Mapping Technique and Process for Tracking and Estimating Evolution of Semantic Networks - In certain embodiments, a computer-implemented method includes accessing first and second data associated with a semantic network, the first data indicating a first plurality of nodes within the semantic network and a first plurality of relationships between the first plurality of nodes at a first time, and the second data indicating a second plurality of nodes within the semantic network and a second plurality of relationships between the second plurality of nodes at a second time. The method further includes generating a first waveform from the first data and a second waveform from the second data. The waveforms indicate an activity level of each of the nodes within the semantic network. The method further includes analyzing the semantic network using the generated first and second waveforms.02-09-2012
20120036092METHOD AND SYSTEM FOR GENERATING A PREDICTION NETWORK - A system and method for creating a network from a number of nodes and edges, where each node is assigned data from at least one data source, the data of a data source being changeable, and wherein the data assigned to a node describe single forecasts from a prediction market, the method comprising structuring the data according to a predefined taxonomy, performing a pattern recognition within data assigned to at least two nodes, whereby the pattern recognition determines and analyzes at least two sequences of patterns of changes, comparing the sequences of patterns and deriving a correlation between the sequences of patterns from the comparison result, wherein the correlation defines the dependency between the nodes; and storing the sequences of patterns and the dependency in a pattern database, whereby the dependency forms an edge between the nodes.02-09-2012
20090182689RULE-BASED DYNAMIC OPERATION EVALUATION - A computer program may involve a dynamic operation, which may specify one of many types of methods based on the conditions of the invocation during runtime, such as the parameters provided to the dynamic operation. The appropriate performance of the dynamic method may be achieved by analyzing the conditions of the invocation according to an evaluation rule set, the rules comprising conditions and an action to be performed if the conditions are satisfied. The evaluation rule set may also be reconfigured upon identifying a satisfied rule to facilitate a faster evaluation of the dynamic operation during a second and subsequent invocations.07-16-2009
20120303558SYSTEMS AND METHODS FOR GENERATING MACHINE LEARNING-BASED CLASSIFIERS FOR DETECTING SPECIFIC CATEGORIES OF SENSITIVE INFORMATION - A computer-implemented method may include (1) identifying a plurality of specific categories of sensitive information to be protected by a DLP system, (2) obtaining a training data set for each specific category of sensitive information that includes a plurality of positive and a plurality of negative examples of the specific category of sensitive information, (3) using machine learning to train, based on an analysis of the training data sets, at least one machine learning-based classifier that is capable of detecting items of data that contain one or more of the plurality of specific categories of sensitive information, and then (4) deploying the machine learning-based classifier within the DLP system to enable the DLP system to detect and protect items of data that contain one or more of the plurality of specific categories of sensitive information in accordance with at least one DLP policy of the DLP system.11-29-2012
20100293116URL AND ANCHOR TEXT ANALYSIS FOR FOCUSED CRAWLING - Systems and methods of URL and anchor text analysis for focused crawling are disclosed. In an exemplary embodiment, a method may include training a focused crawler by: obtaining a training set of at least URL's or anchor text for a website, computing a score for the training set, and extracting a plurality of features of the training set, and computing a score for each of the plurality of features. The features identify key information contained in the website. The method may also include executing a trained focused crawler on other websites.11-18-2010
20100185567SUPERVISION BASED GROUPING OF PATTERNS IN HIERARCHICAL TEMPORAL MEMORY (HTM) - A HTM network that uses supervision signals such as indexes for correct categories of the input patterns to group the co-occurrences detected in the node. In the training mode, the supervised learning node receives the supervision signals in addition to the indexes or distributions from children nodes. The supervision signal is then used to assign the co-occurrences into groups. The groups include unique groups and nonunique groups. The co-occurrences in the unique group appear only when the input data represent certain category but not others. The nonunique groups include patterns that are shared by one or more categories. In an inference mode, the supervised learning node generates distributions over the groups created in the training mode. A top node of the HTM network generates an output based on the distributions generated by the supervised learning node.07-22-2010
20100088257Systems and Methods for Generating Predicates and Assertions - Systems and methods for deriving a predicate by constructing a logic formula from information recorded during test execution, optimizing the logic formula and computing the logical implication of the optimized logic formula. Systems and methods for deriving an assertion from a logical implication by substituting each predicate in the logical implication with corresponding design elements from a hardware design description, inserting the design elements into a target template, inserting a context-sensitive input of the target template based on design elements in the hardware design description and creating an instance name for an instantiation of the target template. Systems and methods for generating a set of clauses that are implied by a disjunctive normal formula of a set of cubes.04-08-2010
20100088255METHOD AND SYSTEM FOR DETERMINING THE ACCURACY OF DNA BASE IDENTIFICATIONS - Embodiments disclosed herein relate to a method and system for determining the accuracy of DNA base identifications, based at least partly on sampling characteristics of subsets within training data sets.04-08-2010
20100138366SYSTEM AND METHOD FOR INFORMATION PROCESSING AND MOTOR CONTROL - The present invention relates to a system and method for information process and motor control using artificially constructed apparatus. More specially, the present invention provides a system and method that can process nature language and other informational input including visual, audio and other sensory inputs and respond intelligently.06-03-2010
20100138365PORTABLE WIRELESS ENABLED DIGITAL MEDIA FRAME - A wireless enabled digital media frame that can communicate over a wireless wide area network (WWAN) is provided. The wireless enabled digital media frame comprises an internal or external WWAN modem (e.g. GPRS/EDGE/UMTS/HSPA/LTE) such that the one or more media is transferred to the media frame using a wireless connection (e.g. 2G/3G/3.5G/4G). The wireless enabled digital media frame displays the received media files on a display screen. Further, broadcast alerts received over the WWAN and/or calculated current signal strength is also displayed on the display screen. Furthermore, setting information is received by the WWAN modem and accordingly applied to modify or update media frame functions.06-03-2010
20080270329SYSTEMS AND METHODS FOR MARTINGALE BOOSTING IN MACHINE LEARNING - Boosting algorithms are provided for accelerated machine learning in the presence of misclassification noise. In an exemplary embodiment, a machine learning method having multiple learning stages is provided. Each learning stage may include partitioning examples into bins, choosing a base classifier for each bin, and assigning an example to a bin by counting the number of positive predictions previously made by the base classifier associated with the bin.10-30-2008
20100082508Method for tagging of a content and a corresponding system - A method generates tag proposals for tagging of a content, wherein the generating of said tag proposals is performed by combining at least two tag proposing procedures in dependence of a work context of a user. The method can be applied with regard to each area where tagging of contents is desired. By use of the method an effective, computing resource saving, and/or flexible tagging is enabled, by which a sufficient number of tags with high quality can be identified.04-01-2010
20130080358ONLINE ASYNCHRONOUS REINFORCEMENT LEARNING FROM CONCURRENT CUSTOMER HISTORIES - In one embodiment, an indication of a Decision Request or an Update Request may be received, where the Update Request is activated independent of user activity. A user state pertaining to at least one user may be received, obtained, accessed or constructed. For the Decision Request, one or more actions may be scored according to one or more value functions associated with a computing device, a policy associated with the computing device may be applied to identify one of the scored actions as a decision, and an indication of the decision may be provided or applied. For the Update Request, the one or more value functions and/or the policy may be updated. An indication of updates to the one or more value functions and/or an indication of updates to the policy may be provided.03-28-2013
20090254500CONTROL SYSTEM FOR NETWORK OF INPUT DEVICES WITH AUTOMATIC AUDIO/VIDEO RECEIVER DETECTION AND CONTROL CONFIGURATION - Apparatus, methods, and systems for centrally and uniformly controlling the operation of a variety of devices, such as communication, consumer electronic, audio-video, analog, digital, 1394, and the like, over a variety of protocols within a network system and, more particularly, a control system and uniform user interface for centrally controlling these devices in a manner that appears seamless and transparent to the user. In a one embodiment, the control system will detect the change of state of an audio output sensor coupled to the audio output port.10-08-2009
20130085970INTELLIGENT INTENT DETECTION FROM SOCIAL NETWORK MESSAGES - An intent engine that automatically detects user intent from messages of a social network (e.g., messages with questions to ask) and outputs intent data. The engine is intelligent in that it can process natural language input such as questions and terms. The user is then directed to an answer page filtered according to the intent data and which proviJoshdes answers related to a question, for example. The intent engine can be designated (e.g., tagged, or “friended”) and then linked into a specialized relationship (e.g., a “friend”). Accordingly, in one example, a URL link is constructed that points to the answer page, with filters configured based on the intent data. The URL is then sent back to the user as a friendly response. When the user selects the link, the user is presented with an answer page that provides answers which match the user intent derived from the user messages.04-04-2013
20100100511CHANGE-POINT DETECTING METHOD AND APPARATUS - To detect a statistical change-point that appears in time-series data with a high accuracy. A first model learning section 102 learns the occurrence probability distribution of time-series data 111 as a first statistical model (for example, a latent Markov model) defined by a finite number of variables including a latent variable. In the subsequent processing, the degree of a temporal change in the probability distribution is computed for each of the probability distribution of the entire first statistical model, its partial probability distribution (the probability distribution of the latent variable and conditional probability distribution contingent on the value of the latent variable), and the probability distribution in which the above plural probability distributions are linearly-combined with a weight. The change-point of the time-series data 111 is detected on the basis of the computed degree of the change.04-22-2010
20100023467RULE LEARNING METHOD, PROGRAM, AND DEVICE - A rule learning method in machine learning includes distributing features to a given number of buckets based on a weight of the features which are correlated with a training example; specifying a feature with a maximum gain value as a rule based on a weight of the training example from each of the buckets; calculating a confidence value of the specified rule based on the weight of the training example; storing the specified rule and the confidence value in a rule data storage unit; updating the weights of the training examples based on the specified rule, the confidence value of the specified rule, data of the training example, and the weight of the training example; and repeating the distributing, the specifying, the calculating, the storing, and the updating, when the rule and the confidence value are to be further generated.01-28-2010
20110173143SYSTEM AND METHOD FOR ANALYZING EXPLORATORY BEHAVIOR - The invention provides a system and method for analyzing a subject's exploratory behavior. The system of the invention includes a tracking device configured to track motion of the subject and to generate a signal indicative of the subject's motion. A CPU analyzes the signal and identifies in the signal sequences of repeated motions, or sequences of sequences of repeated motion, for sequence of repeated motion, the CPU determines for each occurrence of the repeated motion a time at which the occurrence occurred or a time interval during which the occurrence occurred. The CPU then calculates for each occurrence of the repeated motion a value of one or more predetermined parameters of the occurrence of the motion and then calculates a time dependence of the one or more predetermined parameters during the sequence of repeated motion or the sequence of sequences of repeated motion.07-14-2011
20090006283USING A DATA MINING ALGORITHM TO GENERATE FORMAT RULES USED TO VALIDATE DATA SETS - Provided are a method, system, and article of manufacture for using a data mining algorithm to generate format rules used to validate data sets. A data set has a plurality of columns and records providing data for each of the columns. Selection is received of at least one format column for which format rules are to be generated and selection is received of at least one predictor column. A format mask column is generated for each selected format column. For records in the data set, a value in the at least one format column is converted to a format mask representing a format of the value in the format column and storing the format mask in the format mask column in the record for which the format mask was generated. The at least one predictor column and the at least one format mask column are processed to generate at least one format rule. Each format rule specifies a format mask associated with at least one condition in the at least one predictor column.01-01-2009
20090006287Kernel machine approximation - Systems and methods associated with approximating a kernel matrix are described. One method embodiment includes accessing a set of kernel machine training data and then partitioning that data into a set of partitioned data based on a set of partition parameters. The method embodiment may also include creating a partition matrix from the partitioned data, where the partition matrix is to approximate a kernel matrix that may have been created using radial basis functions from the training data.01-01-2009
20090240639Feedback in Group Based Hierarchical Temporal Memory System - A Hierarchical Temporal Memory (HTM) network has at least first nodes and a second node at a higher level than the first nodes. The second node provides an inter-node feedback signal to the first nodes for grouping patterns and sequences (or co-occurrences) in input data received at the first nodes at the first nodes. The second node collects forward signals from the first nodes; and thus, the second node has information about the grouping of the patterns and sequences (or co-occurrences) at the first nodes. The second node provides inter-node feedback signals to the first nodes based on which the first nodes may perform the grouping of the patterns and sequences (or co-occurrences) at the first nodes. Also, a node in a Hierarchical Temporal Memory (HTM) network comprising a co-occurrence detector and a group learner coupled to the co-occurrence detector. The group learner provides an intra-node feedback signal to the co-occurrence detector including information on the grouping of the co-occurrences. The co-occurrence detector may select co-occurrences to be split, merged, retained or discarded based on the intra-node feedback signals.09-24-2009
20120143789CLICK MODEL THAT ACCOUNTS FOR A USER'S INTENT WHEN PLACING A QUIERY IN A SEARCH ENGINE - A method of generating training data for a search engine begins by retrieving log data pertaining to user click behavior. The log data is analyzed based on a click model that includes a parameter pertaining to a user intent bias representing the intent of a user in performing a search in order to determine a relevance of each of a plurality of pages to a query. The relevance of the pages is then converted into training data.06-07-2012
20120143794ANSWER MODEL COMPARISON - This patent application pertains to answer model comparison. One implementation can determine a first frequency at which an individual answer category appears in an individual slot on a query results page when utilizing a first model. The method can ascertain a second frequency at which the individual answer category appears in the individual slot on the query results page when utilizing a second model. The method can calibrate the second model so that the second frequency approaches the first frequency.06-07-2012
20120143796GROUP VARIABLE SELECTION IN SPATIOTEMPORAL MODELING - In response to issues of high dimensionality and sparsity in machine learning, it is proposed to use a multiple output regression modeling module that takes into account information on groups of related predictor features and groups of related regressions, both given as input, and outputs a regression model with selected feature groups. Optionally, the method can be employed as a component in methods of causal influence detection, which are applied on a time series training data set representing the time-evolving content generated by community members, output a model of causal relationships and a ranking of the members according to their influence.06-07-2012
20090276379Using automatically generated decision trees to assist in the process of design and review documentation - An embodiment of this invention is to use automatically generated decision trees to assist in the design and review process. In one embodiment, the decision trees are automatically extracted from data describing a system (in case of design process) or a review artifact (in case of review process). In a further embodiment, the decision trees are then used in the design process, and the order of attributes in the decision tree suggests a new order for writing the design document.11-05-2009
20090276378System and Method for Identifying Document Structure and Associated Metainformation and Facilitating Appropriate Processing - A system and method for processing documents by utilizing the textual content and layout of the documents, including visual indicators, to more efficiently and reliably process the documents across various document types. The system and method identifies visually distinguishable elements within the document, such as section and sub-section boundary indicators, to mark, divide and label the boundaries and content type such that the sections are more clearly identifiable and easily processed. The system and method uses known elements, including section heading types, keywords, section type classifiers, sub-section heading constructs, stop words, and the like to adaptively identify and process a broad range of document types. The system and method continually refines and updates these known elements and allows users to discover and define new elements for further refinement and updating.11-05-2009
20090144209SEQUENCE PREDICTION SYSTEM - The system includes a storage device 06-04-2009
20090287622System and Method for Active Learning/Modeling for Field Specific Data Streams - A system and method for determining whether at least one data point is interesting may be provided. The system may include, among other things, a memory for the at least one data point and a query-by-transduction module configured to assign a plurality of labels to the at least one data point, wherein each label among the plurality of labels corresponds to a respective classification for the at least one data point and wherein each label corresponds to a respective confidence metric that indicates a level of confidence that the respectively corresponding label accurately classifies the at least one data point, analyze the plurality of confidence metrics, and determine whether the at least one data point is interesting based on the analysis.11-19-2009
20090287620SYSTEM AND METHOD FOR OBJECT DETECTION AND CLASSIFICATION WITH MULTIPLE THRESHOLD ADAPTIVE BOOSTING - Systems and methods for classifying a object as belonging to an object class or not belonging to an object class using a boosting method with a plurality of thresholds is disclosed. One embodiment is a method of defining a strong classifier, the method comprising receiving a training set of positive and negative samples, receiving a set of features, associating, for each of a first subset of the set of features, a corresponding feature value with each of a first subset of the training set, associating a corresponding weight with each of a second subset of the training set, iteratively i) determining, for each of a second subset of the set of features, a first threshold value at which a first metric is minimized, ii) determining, for each of a third subset of the set of features, a second threshold value at which a second metric is minimized, iii) determining, for each of a forth subset of the set of features, a number of thresholds, iv) determining, for each of a fifth subset of the set of features, an error value based on the determined number of thresholds, v) determining the feature having the lowest associated error value, and vi) updating the weights, defining a strong classifier based on the features having the lowest error value at a plurality of iterations, and classifying a sample as either belonging to an object class or not belonging to an object class based on the strong classifier.11-19-2009
20090089227AUTOMATED RECOMMENDATIONS FROM SIMULATION - An industrial controller simulation system is provided. The system includes a simulation component that enables modeling of an industrial controller system. A suggestion component offers automated recommendations in accordance with the modeling of the industrial controller system.04-02-2009
20080208775Method and Apparatus for Generation of a Sequence of Elements - A method and apparatus for automatically generating a target sequence of a plurality of elements selected in accordance with a plurality of user-defined constraints such as a play list of songs. The apparatus comprises a user interface (08-28-2008
20080208776METHOD AND APPARATUS FOR LEARNING BEHAVIOR IN SOFTWARE ROBOT - Disclosed is a method and apparatus for learning behavior in a software robot. The method includes detecting a kind of an object in cyberspace related to a kind of presently manifested action, and a kind and the variation of at least one state among percept states or emotional states preset so as to change in relation to the kind of the action; finding episodes respectively corresponding to each of one or more objects in the cyberspace, each of one or more emotional states and each of one or more percept states, respectively defined in the software robot, a kind of an object in cyberspace related to the detected kind of the action among multiple episodes for responding a combination of kinds of respective one or more actions and for storing variation related to each state, and a kind of at least one state among percept states or emotional states preset so as to change in relation to the kind of the action; using variation stored in response to the found episode and variation generated in response to the manifested action, and calculating a representative variation; and storing the representative variation as a variation of the found episode.08-28-2008
20110173142APPARATUS AND METHODS FOR CLASSIFYING SENDERS OF UNSOLICITED BULK EMAILS - Disclosed are methods and apparatus for facilitating the filtering of unsolicited bulk electronic mail (email) sent from spammers. A plurality of recipient patterns for a plurality of emails from known spammers is logged. A plurality of recipient patterns for a plurality of emails from known non-spammers is also logged. A probabilistic model for predicting whether an unknown sender identity is a spammer is generated or modified based on the logged recipient patterns for the emails from known spammers and known non-spammers.07-14-2011
20090276381 SOCIAL KNOWLEDGE SYSTEM CONTENT QUALITY - Techniques for automatically scoring submissions to an online question-and-answer submission system are disclosed. According to one such technique, an initial set of user submissions are scored by human operators and/or automated algorithmic mechanisms. The submissions and their accompanying scores are provided as training data to an automated machine learning mechanism. The machine learning mechanism processes the training data and automatically detects patterns in the provided submissions. The machine learning mechanism automatically correlates these patterns with the scores assigned to the submissions that match those patterns. As a result, the machine learning mechanism is trained. Thereafter, the machine learning mechanism processes unscored submissions. The machine learning mechanism automatically identifies, from among the patterns that the machine learning mechanism has already detected, one or more patterns that these submissions match. The machine learning mechanism automatically scores these submissions based on the matching patterns and the scores that are associated with those patterns.11-05-2009
20090276380COMPUTER-AIDED NATURAL LANGUAGE ANNOTATION - The present invention uses a natural language understanding system that is currently being trained to assist in annotating training data for training that natural language understanding system. Unannotated training data is provided to the system and the system proposes annotations to the training data. The user is offered an opportunity to confirm or correct the proposed annotations, and the system is trained with the corrected or verified annotations.11-05-2009
20090276377NETWORK DATA MINING TO DETERMINE USER INTEREST - Mining information from network data traffic to determine interests of online network users is provided herein. A data packet received at a network interface device can be accessed and inspected at line rate speeds. Source or addressing information in the data packet can be extracted to identify an initiating and/or receiving device. The packet can be inspected to identify occurrences of keywords or data features related with one or more subject matters. A vector can be defined for a network device that indicates a relative rank of interest in various subject matters. Furthermore, statistical analysis can be implemented on data stored in one or more interest vectors to determine information pertinent to network user interests. The information can facilitate providing value-added products or services to network users.11-05-2009
20090281969Decision Tree Representation of a Function - An arbitrary function may be represented as an optimized decision tree. The decision tree may be calculated, pruned, and factored to create a highly optimized set of equations, much of which may be represented by simple circuits and little, if any, complex processing. A circuit design system may automate the decision tree generation, optimization, and circuit generation for an arbitrary function. The circuits may be used for processing digital signals, such as soft decoding and other processes, among other uses.11-12-2009
20090281970AUTOMATED TAGGING OF DOCUMENTS - An automated technique for tagging documents includes using a semantic tagger to generate an annotation that associates a standard tag with a first text fragment of the user-defined document, wherein the tagger is trained on a standard document annotated with a standard tag, associating the first user-defined tag with a second text fragment of the user-defined document in response to the second text fragment matching a value associated with the first user-defined tag, and establishing a mapping between the standard tag and the first user-defined tag in response to existence of a requisite correlation between the standard tag and the user-defined tag. The technique may further include selecting from the user-defined document a tagged text fragment that is associated with a second user-defined tag, and providing the tagged text fragment and a standard tag associated by the mapping with the second user-defined tag to the tagger as additional training input.11-12-2009
20090287621Forward feature selection for support vector machines - In one embodiment, the present invention includes a method for training a Support Vector Machine (SVM) on a subset of features (d′) of a feature set having (d) features of a plurality of training instances to obtain a weight per instance, approximating a quality for the d features of the feature set using the weight per instance, ranking the d features of the feature set based on the approximated quality, and selecting a subset (q) of the features of the feature set based on the ranked approximated quality. Other embodiments are described and claimed.11-19-2009
20120143795CROSS-TRACE SCALABLE ISSUE DETECTION AND CLUSTERING - Techniques and systems for cross-trace scalable issue detection and clustering that scale-up trace analysis for issue detection and root-cause clustering using a machine learning based approach are described herein. These techniques enable a scalable performance analysis framework for computing devices addressing issue detection, which is designed as a multiple scale feature for learning based issue detection, and root cause clustering. In various embodiments the techniques employ a cross-trace similarity model, which is defined to hierarchically cluster problems detected in the learning based issue detection via butterflies of trigram stacks. The performance analysis framework is scalable to manage millions of traces, which include high problem complexity.06-07-2012
20110270787VERIFICATION SUPPORT COMPUTER PRODUCT, APPARATUS, AND METHOD - A non-transitory, computer-readable recording medium stores therein a verification support program that causes a computer to execute identifying from a finite state machine model related to a circuit-under-test, an input count of transitions to a transition-end state and an output count of transitions from the transition-end state; determining the transition-end state to be a record/restore subject, if the identified output transition>the identified input transition count; embedding record-instruction information causing the record/restore subject to be recorded to a database, if a first element causing transition to the record/restore subject is included in a first test scenario that is in a test scenario group related to the circuit-under-test; and embedding restore-instruction information causing the record-restore subject to be restored from the database, if a second element causing transition to the record-restore subject is included in a series of elements making up a second test scenario that is in the test scenario group.11-03-2011
20100082506Active Electronic Medical Record Based Support System Using Learning Machines - A data processing technique is provided. In one embodiment, a computer-implemented method includes receiving image data from an imaging system and organizing the image data into multiple objects of interest. The method may also include identifying source-invariant features of the multiple objects of interest and classifying the multiple objects of interest via a learning algorithm into categories based, at least in part, on the identified source-invariant features. Further, the method may include outputting a report based at least in part on data derived from the classification of one or more of the multiple objects of interest. Additional methods, systems, and devices are also disclosed.04-01-2010
20100145891GENERATING EXTENDED DATA FROM A PATTERN-RECOGNITION MODEL FOR A COMPUTER SYSTEM - Some embodiments of the present invention provide a system that generates extended data for a pattern-recognition model used in electronic prognostication for a computer system. During operation the system determines, for each sensor in a set of sensors, a regression coefficient between training data from the sensor and training data from each of the other sensors in the set of sensors. Next, for each sensor in the set of sensors, the system stretches the training data from each of the other sensors by a predetermined amount, and generates extended data for the sensor based on the stretched training data for each of the other sensors and the regression coefficients between training data from the sensor and training data from each of the other sensors.06-10-2010
20100145896COMPOUND PROPERTY PREDICTION APPARATUS, PROPERTY PREDICTION METHOD, AND PROGRAM FOR IMPLEMENTING THE METHOD - A compound property prediction apparatus includes a training sample library (06-10-2010
20080243730Training a machine learning system to determine photoresist parameters - To train a machine learning system, a set of different values of one or more photoresist parameters, which characterize behavior of photoresist when the photoresist undergoes processing steps in a wafer application, is obtained. A set of diffraction signals is obtained using the set of different values of the one or more photoresist parameters. The machine learning system is trained using the set of measured diffraction signals as inputs to the machine learning system and the set of different values of the one or more photoresist parameters as expected outputs of the machine learning system.10-02-2008
20080243729LEVERAGING USER-TO-USER INTERACTIONS IN A KNOWLEDGEBASE USING A FORUM INTERFACE - Systems and methods provide a self-learning knowledgebase in which the ranking and/or order of topic and thread items may be dynamically and automatically adjusted based on self-learning by the knowledgebase. The knowledgebase includes threaded conversations comprising thread topics and thread items within the thread topics. Lists of thread topics and lists of thread items are ordered lists. The order of a thread topic or thread item in an ordered list may be modified based on self-learning activities performed by an information server maintaining the knowledge base. A thread topic or thread item may be moved higher in the list based on requests to view the thread topic or thread item. Further, the order that a thread item appears in an order list may be modified based on a number of responses posted for the thread item.10-02-2008
20080275827Parallelization of Bayesian Network Structure Learning - A master computing node directs parallel structure learning with intelligent computational task distribution. The master computing node may determine what families are to be used to score neighbors in a neighbor scoring process, and determine if the families have scores in a score cache. Families to be scored for the score cache may be marked and distributed for calculation among nodes in the computing cluster. The score cache may be updated to include the scored families, and the cluster synchronized with the score cache data.11-06-2008
20090055332METHOD OF GENERATING ASSOCIATION RULES FROM DATA STREAM AND DATA MINING SYSTEM - Disclosed is a method and data mining system for generating association rules from a data stream. An embodiment of the invention provides a method of generating association rules from a data stream, which is a non-limited data set composed of transactions continuously generated. The method includes: when itemsets included in the generated transactions and the counts of the itemsets are managed using a prefix tree and each node of the prefix tree has information on the count of a specific itemset corresponding to the node and a specific item, updating the information of a node corresponding to the itemset or adding a new node on the basis of the itemset included in the generated transaction and the count of the itemset; comparing the support of the itemset corresponding to each of the nodes of the prefix tree with a minimum support, which is a predetermined threshold value, to select frequent itemsets; and visiting all or some of the nodes corresponding to the selected frequent itemsets, and generating the association rule on the basis of the information of each of the visited nodes.02-26-2009
20130218818CROSS CHANNEL OPTIMIZATION SYSTEMS AND METHODS - The inventive subject matter is generally directed to a cross channel optimization system, methods, and related software which provide for the conducting of experiments and/or optimization of digital content across a plurality of external content systems to user of the external content systems.08-22-2013
20130218815METHODS AND SYSTEMS FOR FEATURE EXTRACTION - A method and system for extracting feature utilizing an AHaH module (Anti-Hebbian and Hebbian). A sparse input data stream can be presented to a synaptic matrix of a collection of AHaH nodes associated with the AHaH module. The AHaH module operates an AHaH plasticity rule via an evaluate phase and a feedback phase cycle. A bias input line can be modulated such that a bias weight do not receive a Hebbian portion of the weight update during the feedback phase in order to prevent occupation of a null state. The input space can be bifurcated when the AHaH nodes fall randomly into an attractor state. The output of the AHaH module that forms a stable bit pattern can then be provided as an input to a content-addressable memory for generating a maximally efficient binary label.08-22-2013
20130218816APPARATUS AND METHOD FOR PROCESSING SENSOR DATA IN SENSOR NETWORK - In a sensor network, a sensor data processing apparatus generates a feature vector identifier table by classifying feature vector identifiers of a plurality of situation information determination reference data to be a reference of situation determination according to a sensor type index and a feature vector identifier set index of the plurality of situation information reference data. When the sensor data processing apparatus receives sensor data, the sensor data processing apparatus generates a feature vector identifier of the sensor data and extracts a sensor type index and a feature vector identifier set index of a feature vector identifier most similar to the feature vector identifier of sensor data with reference to a feature vector identifier table, and generates situation recognition information using the extracted sensor type index and feature vector identifier set index.08-22-2013
20080228676Computing device, method of controlling the computing device, and computer readable medium recording a program - A computing device stores a Bayesian network (09-18-2008
20080288425Methods and Apparatus for Reasoning About Information Fusion Approaches - Methods and apparatus for reasoning about information fusion approaches are described according to some aspects. In one aspect, a method for reasoning about information fusion approaches comprises selecting one or more information fusion approaches for evaluation, and applying the approaches to two or more sets of data. The selection of information fusion approaches can be based on a taxonomy that classifies information fusion algorithms. The impact of each of the applied information fusion approaches can then be conveyed to a user, wherein the conveyed impact is based on one or more impact measures.11-20-2008
20100280978System and method for utility usage, monitoring and management - Systems and methods are provided for collecting waveform data for a plurality of appliances that may be found in a residential or commercial setting using multi-port outlet monitoring devices to obtain power consumption profiles that indicate power consumption on a per-appliance and/or per-location basis and/or per user basis. The plurality of appliances is reliably identified from the power consumption profiles. In accordance with a method embodiment, waveform data transmitted from an unknown appliance is independently metered via a multi-port monitoring device over an elapsed time period. The metered waveform data is wirelessly transmitted from the multi-port monitoring device to a co-located system controller which constructs an appliance signature. The process may be repeated to generate multiple appliance signatures. The one or more appliance signatures are compared to a database of pre-stored canonical signatures to determine if there is a match to identify the appliance.11-04-2010
20090265291Information Processing Device and Method, and Program - An information processing device includes: a candidate generating unit employing a user evaluation matrix of evaluation values indicating evaluations as to multiple contents for multiple users to generate multiple estimated expression candidates which are candidates of an estimated expression employed for estimating an evaluation as to a content of a user; an estimation results computing unit computing the user evaluation matrix by the respective estimated expression candidates to generate an estimation result configured of a predictive evaluation value which is the estimation value of an evaluation value; and an estimated expression selecting unit, in a case where several estimation results are employed, and several estimated expression candidates are employed as estimated expressions, obtaining linear combination coefficients employed for obtaining a final estimation result, and selecting an estimated expression candidate and linear combination coefficient having the highest evaluation as the estimated expression and linear combination coefficient of the next generation.10-22-2009
20090265290OPTIMIZING RANKING FUNCTIONS USING CLICK DATA - A system for optimizing machine-learned ranking functions based on click data. The system determines the weighting for each feature of a plurality of features according to a learning model based on the click data. The system selects an element from a plurality of elements for display on a web page based on the weighting of each feature of the plurality of features. The system may rank the items to form a list on the web page based on the weighted features in order of inferred relevance according to the online learning model.10-22-2009
20080313110METHOD AND SYSTEM FOR SELF-CALIBRATING PROJECT ESTIMATION MODELS FOR PACKAGED SOFTWARE APPLICATIONS - An estimation system for deriving multi-dimensional project plans for implementing packaged software applications with self-calibration and refinement of project estimation models, the system includes: a view layer configured to act as a user interface for user inputs and system outputs; a model and control layer configured to implement rules based on a series of estimation and implementation models, and to perform self-calibration and refinement of project estimation models for multi-dimensional project plans; an estimation knowledge base layer configured to hold and derive the series of estimation and implementation models; and wherein the system for self-calibration and refinement of project estimation models for multi-dimensional project plans for implementing packaged software applications is carried out over networks comprising: the Internet, intranets, local area networks (LAN), and wireless local area networks (WLAN).12-18-2008
20080313112LEARNING MACHINE THAT CONSIDERS GLOBAL STRUCTURE OF DATA - A new machine learning technique is herein disclosed which generalizes the support vector machine framework. A separating hyperplane in a separating space is optimized in accordance with generalized constraints which dependent upon the clustering of the input vectors in the dataset.12-18-2008
20080313111LARGE SCALE ITEM REPRESENTATION MATCHING - A two-phase process quickly and accurately identifies representations of the same items within a collection of item representations. In the first phase, referred to as a “blocking phase,” frequency information indicating the frequency with which terms appear within the collection of item representations is used to quickly identify “candidate pairs” (i.e., pairs of item representations that have a relatively high probability of matching). The blocking phase results in a reduced subset of the data for further analysis during the second phase. In the second phase, referred to as a “matching phase,” the candidate pairs are analyzed using fuzzy matching functions to accurately identify “matching pairs” (i.e., representations of the same items).12-18-2008
20090319449PROVIDING CONTEXT FOR WEB ARTICLES - An overwhelming number of articles are available everyday via the internet. Unfortunately, it is impossible to peruse more than a handful, and it is difficult to ascertain an article's social context. The techniques disclosed herein address this problem by harnessing implicit and explicit contextual information from social media. By extracting text surrounding a hyperlink to an article in a post and assessing the article as a function of content surrounding the hyperlink, an article's social context is determined and presented. Additionally, articles that are sufficiently similar in content may be grouped to establish a many-to-one relationship between posts and an article, creating a more accurate assessment.12-24-2009
20120296856Recognition Techniques to Enhance Automation In a Computing Environment - Systems and methods for detecting end of a transaction in a computing environment are provided. The method comprises determining a target area in a graphical user environment displayed on a display screen, wherein a change is expected to occur when end of a transaction is reached; masking the target area at least partially to remove content included in the target area that is present before or after the transaction was initiated; monitoring the target area for change in content; and detecting the end of the transaction when the content of the target area has changed.11-22-2012
20090313189METHOD, SYSTEM AND APPARATUS FOR ASSEMBLING AND USING BIOLOGICAL KNOWLEDGE - Disclosed are methods, systems and apparatus for constructing assemblies of biological knowledge constituting a biological knowledge base, and for subsetting and transforming life sciences-related data and information into biological models to facilitate computation and electronic reasoning on biological information. A subset of data is extracted from a global knowledge base or repository to reconstruct a more specialized sub-knowledge base or assembly designed specifically for the purpose at hand. Assemblies generated by the invention permit selection and rational organization of seemingly diverse data into a model of any selected biological system, as defined by any desired biological criteria. These assemblies can be mined easily and can be logically reasoned with great productivity and efficiency.12-17-2009
20080208778CONTROLLING A NON-LINEAR PROCESS - System and method for modeling a nonlinear process. A combined model for predictive optimization or control of a nonlinear process includes a nonlinear approximator, coupled to a parameterized dynamic or static model, operable to model the nonlinear process. The nonlinear approximator receives process inputs, and generates parameters for the parameterized dynamic model. The parameterized dynamic model receives the parameters and process inputs, and generates predicted process outputs based on the parameters and process inputs, where the predicted process outputs are useable to analyze and/or control the nonlinear process. The combined model may be trained in an integrated manner, e.g., substantially concurrently, by identifying process inputs and outputs (I/O), collecting data for process I/O, determining constraints on model behavior from prior knowledge, formulating an optimization problem, executing an optimization algorithm to determine model parameters subject to the determined constraints, and verifying the compliance of the model with the constraints.08-28-2008
20080208773Multi-core stochastic discrimination - In some embodiments, multi-core stochastic discrimination is generally presented. In this regard, a method is introduced comprising providing random regions of a feature space to parallel cores, testing each random region for enrichment in parallel, recording coverage for each data point in each enriched random region in parallel, and calculating an overall average coverage for each data point among the enriched random regions. Other embodiments are also disclosed and claimed.08-28-2008
20080235163SYSTEM AND METHOD FOR ONLINE DUPLICATE DETECTION AND ELIMINATION IN A WEB CRAWLER - As part of the normal crawling process, a crawler parses a page and computes a de-tagged hash, called a fingerprint, of the page content. A lookup structure consisting of the host hash (hash of the host portion of the URL) and the fingerprint of the page is maintained. Before the crawler writes a page to a store, this lookup structure is consulted. If the lookup structure already contains the tuple (i.e., host hash and fingerprint), then the page is not written to the store. Thus, a lot of duplicates are eliminated at the crawler itself, saving CPU and disk cycles which would otherwise be needed during current duplicate elimination processes.09-25-2008
20090157573System And Method For Grading Electricity Distribution Network Feeders Susceptible To Impending Failure - A machine learning system creates failure-susceptibility rankings for feeder cables in a utility's electrical distribution system. The machine learning system employs martingale boosting algorithms and Support Vector Machine (SVM) algorithms to generate a feeder failure prediction model, which is trained on static and dynamic feeder attribute data. Feeders are dynamically ranked by failure susceptibility and the rankings displayed to utility operators and engineers so that they can proactively service the distribution system to prevent local power outages. The feeder rankings may be used to redirect power flows and to prioritize repairs. A feedback loop is established to evaluate the responses of the electrical distribution system to field actions taken to optimize preventive maintenance programs.06-18-2009
20100138371INFORMATION PROCESSING APPARATUS AND UPDATE INFORMATION OBTAINMENT METHOD - An information processing apparatus includes an obtainment section that obtains, via a network, information updated at a distribution origin on the network; and a determination section that determines an obtainment rule relating to a timing of the obtainment by the obtainment section of update information for the distribution origin, wherein the obtainment section obtains update information based on a predetermined learning rule for a first distribution origin for which obtainment rule has not been determined by the determination section, the determination section determines an obtainment rule for the first distribution origin based on a result of the obtainment by the obtainment section of update information from the distribution origin based on the learning rule, and the obtainment section, in response to the determination of the obtainment rule by the determination section, obtains update information from the first distribution origin based on the obtainment rule.06-03-2010
20100138368METHODS AND SYSTEMS FOR SELF-IMPROVING REASONING TOOLS - Implementations that integrate data-driven modeling and knowledge into self-improving reasoning systems and processes are described. For example, an implementation of a method may include determining at least one recommended action using a reasoning component having a data-driven modeling portion and a knowledge-based portion. Such determining includes integrating one or more determination aspects determined by the data-driven modeling portion, and one or more additional determination aspects determined by the knowledge-based portion.06-03-2010
20130191311OPTIMIZING ELECTRONIC COMMUNICATION CHANNELS - A method, computer program product, and system for electronic communication is described. A first unified telephony session is selected. A first arbitrator associated with the first session is selected. A first set of participants associated with the first session is selected. The first arbitrator is directed to act as a proxy connection for a first channel associated with the first set of participants.07-25-2013
20130191310PREDICTION MODEL REFINEMENT FOR INFORMATION RETRIEVAL SYSTEM - A learning system refines a prediction model that determines the effectiveness of a search engine in achieving a goal of a search. Search goal achievements are estimated for sequences of user actions in an unlabeled data set using the prediction model, which is based on a mixture model and values for parameters of the mixture model. The values of the parameters are redefined based at least on the search goal achievement estimates of the unlabeled set. The prediction model is stored in accordance with the mixture model and the redefined values.07-25-2013
20100005042Support vector regression for censored data - A method of producing a model for use in predicting time to an event includes obtaining multi-dimensional, non-linear vectors of information indicative of status of multiple test subjects, at least one of the vectors being right-censored, lacking an indication of a time of occurrence of the event with respect to the corresponding test subject, and performing regression using the vectors of information to produce a kernel-based model to provide an output value related to a prediction of time to the event based upon at least some of the information contained in the vectors of information, where for each vector comprising right-censored data, a censored-data penalty function is used to affect the regression, the censored-data penalty function being different than a non-censored-data penalty function used for each vector comprising non-censored data.01-07-2010
20100005041MACHINE LEARNING BASED VOLUME DIAGNOSIS OF SEMICONDUCTOR CHIPS - A system and method for integrated circuit diagnosis includes partitioning an integrated circuit design into sub-regions according to a structure of the integrated circuit design. A decision function is generated for a sub-region by training a machine learning tool. A sequence of test patterns is applied to a device under test (DUT) to determine responses. If the DUT fails, all the decision functions are evaluated with the errors produced by the DUT. A sub-region whose decision function yielded a highest value is selected to find a defect sub-region in the DUT.01-07-2010
20100005040FORECASTING ASSOCIATION RULES ACROSS USER ENGAGEMENT LEVELS - A method of determining one or more association rules includes: specifying site-sequence values for users, wherein each user is identified with one of a plurality of engagement levels, and the site-sequence values indicate a sequence from a first site to a second site for at least one user identified with a corresponding engagement level; determining cumulative site-sequence values from the site-sequence values for combinations of pairs of sites and distinct engagement levels; determining likelihood values from the cumulative site-sequence values, wherein the likelihood values characterize probabilities for sequences between sites at distinct engagement levels; determining one or more association rules for pairs of sites from one or more corresponding likelihood values at one or more engagement levels, wherein each association rule indicates a sequential association between a corresponding pair of sites; determining one or more confidence values for the one or more association rules by calculating one or more variations of the likelihood values across the engagement levels; and saving one or more values for the one or more associations rules (e.g., likelihood values or confidence values).01-07-2010
20110208680ASSISTING WITH UPDATING A MODEL FOR DIAGNOSING FAILURES IN A SYSTEM - The method includes obtaining system model data representing a set of failures in a system including a plurality of components, a set of symptoms and relationships between at least some of the failures and symptoms. The system model data is used to create a Bayesian Network. Failure cases data is also obtained, where each failure case describes the presence/absence of at least one of the symptoms and the presence/absence of at least one of the failures. A learning operation on the Bayesian Network using the failure cases data is then performed and the contribution made by at least some of the failure cases to updating the parameters of the Bayesian Network during the learning operation is assessed. Information representing the assessed contribution of the at least some failure cases is displayed.08-25-2011
20110208679TROUBLE PATTERN CREATING PROGRAM AND TROUBLE PATTERN CREATING APPARATUS - A computer readable, non-transitory medium has stored therein a trouble pattern creating program. The program causes a computer to execute: (a) extracting, from a plurality of log messages that are output from an information system having a plurality of configuration items and that are output in a predetermined period of time, configuration items that output the log messages; (b) calculating a degree of relationship between the configuration items extracted in the (a) extracting; (c) executing learning of the rate of the number of occurrences of troubles in the information system in the number of times the log messages are output, the learning is executed by a number of times corresponding to the degree of relationship calculated in the (b) calculating; and (d) creating, in accordance with a result of the learning in the (c) executing, a trouble pattern message that is output when a trouble occurs.08-25-2011
20110208678MECHANICAL SHOCK FEATURE EXTRACTION FOR OVERSTRESS EVENT REGISTRATION - An electronic system includes an accelerometer. A method for excessive mechanical shock feature extraction for overstress event registration and cumulative tracking includes obtaining a sample from the accelerometer. Feature extraction is performed on the sample using empirical mode decomposition (EMD) to produce a plurality of modes. A pattern classifier is utilized for processing the plurality of modes to determine if the sample classifies as a shock event. If the sample classifies as a shock event, a shock event counter is incremented. If the shock event counter reaches a specified count, an indication to a user is generated.08-25-2011
20090132442Method and Apparatus for Determining Decision Points for Streaming Conversational Data - A method for determining a decision point in real-time for a data stream from a conversation includes receiving streaming conversational data; and determining when to classify the streaming conversational data, using a measure of certainty, by performing certainty calculations at a plurality of time instances during the conversation and by selecting a decision point in response to the certainty calculations, the decision point not being based on a fixed window of conversational data but being based on accumulated conversational data available at different ones of the plurality of time instances. Systems and computer program products are also provided.05-21-2009
20110060707INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing device includes: a hierarchy processing unit for generating a unit to connect the unit in a hierarchical structure, the unit including an input control unit for performing input control for storing an observed value, and outputting the time series of the observed value as input data to be given to a learning model having an HMM (Hidden Markov Model) as a minimum component module, a model processing unit for performing processing using the learning model, including a module learning unit for obtaining likelihood of the input data being observed with the module, to determine one module of the learning model, or new module to be an object module having HMM parameters to be updated, and to perform module learning processing for updating the HMM parameters, and a recognizing unit for recognizing the input data using the learning model, and an output control unit for performing output control.03-10-2011
20110060706INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing device comprising: a likelihood calculating unit configured to take the time series of an observed value to be successively supplied as learned data to be used for learning, and with regard to each module making up a learning model having an HMM (Hidden Markov Model) as a module which is the minimum component, to obtain likelihood that the learned data may be observed at the module; an object module determining unit configured to determine, based on the likelihood, a single module of the learning model, or a new module to be an object module that is an object module having an HMM parameter to be updated; and an updating unit configured to perform learning for updating the HMM parameter of the object module using the learned data.03-10-2011
20090187516SEARCH SUMMARY RESULT EVALUATION MODEL METHODS AND SYSTEMS - Methods and systems are provided herein for establishing and/or using an evaluation model that is adapted to determine a model judgment value based, at least in part, on measured summary feature values associated with a search result summary. The evaluation model may be established through a learning process based, at least in part, on human judgment values associated with a set of search result summaries.07-23-2009
20090164394AUTOMATED CREATIVE ASSISTANCE - The claimed subject matter provides a system and/or a method that facilitates generating a suggestion for a creative work. An interface component can receive a portion of a creative work. A muse component can evaluate the portion of the creative work utilizing a machine learning technique, wherein the muse component can identify a portion of suggested content based upon the evaluation. A palette can be populated with the portion of suggested content, wherein the portion of suggested content is a subset of content that is available for incorporation into the creative work.06-25-2009
20090132444CONSTRAINED LINE SEARCH OPTIMIZATION FOR DISCRIMINATIVE TRAINING OF HMMS - An exemplary method for optimizing a continuous density hidden Markov model (CDHMM) includes imposing a constraint for discriminative training, approximating an objective function as a smooth function of CDHMM parameters and performing a constrained line search on the smoothed function to optimize values of the CDHMM parameters. Various other methods, devices and systems are disclosed.05-21-2009
20090187519Learning Device Interaction Rules - Devices and methods are disclosed for establishing interaction among electronic devices of an environment. The device has a transmitter, receiver, memory for storing interaction rules, and a processor for learning the interaction rules in association with the transmitter, receiver, and other devices of the environment. The device also includes components for performing the device specific functions and a state sensor for determining the logical or physical state of the device. Methods involve observing at one or more devices change of state activity among the plurality of devices through receiving a change of state message that is transmitted to the one or more devices. A set of rules are learned at the one or more devices based upon observing the change of state activity. The learned set of rules are then applied at the one or more devices to automatically control changes of state of devices within the plurality of devices.07-23-2009
20090187517MODIFICATION OF RELATIONAL MODELS - Described herein is a system that facilitates modifying a relational model. The system includes a first model component that is a relational model that includes a plurality of atoms. The system further includes a modifier component that automatically assigns values to a plurality of atoms in the relational model by clustering atoms of the relational model to create a second model component, wherein the second model component is a relational model.07-23-2009
20090006288Information Processing Apparatus, Information Processing Method, and Program - An information processing apparatus learning a preference of a user for a content item includes acquiring means for acquiring an operation or expression of the user for a certain content item as feedback information; training data generating means for generating training data for the preference learning from the feedback information acquired by the acquiring means; and learning means for learning the preference of the user and how to attach a meaning to the feedback information in association with the training data by using multiple pieces of training data generated by the training data generating means.01-01-2009
20090006289Hierarchical Temporal Memory System with Enhanced Inference Capability - A node, a computer program storage medium, and a method for a hierarchical temporal memory (HTM) network where at least one of its nodes generates a top-down message and sends the top-down message to one or more children nodes in the HTM network. The first top-down message represents information about the state of a node and functions as feedback information from a current node to its child node. The node may also maintain history of the input patterns or co-occurrences so that temporal relationships between input patterns or co-occurrences may be taken into account in an inference stage. By providing the top-town message and maintaining history of previous input patterns, the HTM network may, among others, (i) perform more accurate inference based on temporal history, (ii) make predictions, (iii) discriminate between spatial co-occurrences with different temporal histories, (iv) detect “surprising” temporal patterns, (v) generate examples from a category, and (vi) fill in missing or occluded data.01-01-2009
20090006282USING A DATA MINING ALGORITHM TO GENERATE RULES USED TO VALIDATE A SELECTED REGION OF A PREDICTED COLUMN - Provided are an article of manufacture, system, and method for using a data mining algorithm to generate rules used to validate a selected region of a predicted column. A data set has a plurality of columns and records providing data for each of the columns. Selection is received of at least one predicted column for which rules are to be generated and at least one region of the selected at least one predicted column, wherein each region specifies data positions in the column. The data set is processed to determine association relationships among data in at least one predictor column and subsequences in the selected at least one region of the at least one predicted column. At least one rule is generated from the relationships specifying a condition involving at least one predictor column that predicts at least one value in the selected region of the at least one predicted column.01-01-2009
20090006285CONTENT-BASED TAGGING OF RSS FEEDS AND E-MAIL - Providing for automated generation of tags (e.g., metadata descriptors) for items of e-mail or syndication formatted communication is described herein. By way of example, a system can include a filtering component that can generate one or more tags based on information relevant to content of the communication, a sender, or recipient, or combinations thereof. In addition, such tags can be automatically attached to a received item, or a presentation component can furnish the tags to a recipient (e.g., by way of a communication device user interface) for selection, whereby selected tags are associated with the item of communication. Accordingly, the subject innovation provides for improved classification and description of items of communication by automatic generation of descriptive and/or representative tags associated therewith.01-01-2009
20090006284FORECASTING TIME-INDEPENDENT SEARCH QUERIES - Techniques for analyzing and modeling the frequency of queries are provided by a query analysis system. A query analysis system analyzes frequencies of a query over time to determine whether the query is time-dependent or time-independent. The query analysis system forecasts the frequency of time-dependent queries based on their periodicities. The query analysis system forecasts the frequency of time-independent queries based on causal relationships with other queries. To forecast the frequency of time-independent queries, the query analysis system analyzes the frequency of a query over time to identify significant increases in the frequency, which are referred to as “query events” or “events.” The query analysis system forecasts frequencies of time-independent queries based on queries with events that tend to causally precede events of the query to be forecasted.01-01-2009
20090187515QUERY SUGGESTION GENERATION - Described herein is a system that facilitates assigning indications of usefulness to query suggestions. The system includes a query suggestion generator component that receives a query and generates a query suggestion based at least in part upon the received query. A model component outputs an indication of usefulness with respect to the query suggestion, wherein the model component is a machine-learned model of user behavior with respect to query suggestions.07-23-2009
20130218819SYSTEM AND METHOD TO ESTIMATE REGION OF TISSUE ACTIVATION - A computer-implemented method for determining the volume of activation of neural tissue. In one embodiment, the method uses one or more parametric equations that define a volume of activation, wherein the parameters for the one or more parametric equations are given as a function of an input vector that includes stimulation parameters. After receiving input data that includes values for the stimulation parameters and defining the input vector using the input data, the input vector is applied to the function to obtain the parameters for the one or more parametric equations. The parametric equation is solved to obtain a calculated volume of activation.08-22-2013
20130218814METHOD AND SYSTEM FOR THE DYNAMIC ALLOCATION OF RESOURCES BASED ON FAIRNESS, THROUGHPUT, AND USER BEHAVIOR MEASUREMENT - A system and method for the dynamic allocation of resources based on fairness, throughput, and user behavior measurement. A resource allocation decision can be made based on an index value computed by a selection index function, A fairness coefficient and a throughput coefficient, which represents the significance of fairness and throughput can be computed utilizing a reinforcement learning algorithm and the degree of fairness and throughput coefficient can be varied while allocating resources. A user behavior coefficient with respect to a user can be computed to determine the degree of cooperativeness of the user with other users and the value of user behavior coefficient can be updated each time it interacts with the system.08-22-2013
20090327175PHARMACOKINETIC MODELING OF MYCOPHENOLIC ACID - A method of providing a pharmacokinetic model to provide optimize pharmacokinetic data associated with administering a drug to a patient and a method of optimising pharmacokinetic data associated with administering a drug to a patient, data processing apparatus, recording medium and a pharmacokinetic model are disclosed.12-31-2009
20090327174TASK HISTORY USER INTERFACE USING A CLUSTERING ALGORITHM - The aspects of the disclosed embodiments include clustering a set of discrete user interface states into groups; presenting the groups on a display of a device; and enabling selection of any state within a presented group, wherein selection of a state returns the user interface to the selected state.12-31-2009
20090144210METHOD AND APPARATUS FOR DETERMINING THE VARIABLE DEPENDENCY - A method and an apparatus for determining variable dependency are disclosed. In the present invention, a variable dependency is determined in advance arbitrarily; partial variables are selected from the current variable dependency, and a legitimate superior variable set is re-selected for each of the partial variables, and the new variable dependency is stored only if it meets the criterion of acceptance; when the termination criterion for establishing variable dependency is met, the optimal variable dependency is determined from all variable dependencies. Because the existing variable dependency is not taken as a reference when the new variable dependency is created, the new variable dependency is not misled by the existing variable dependency, and the time for finding the globally optimal variable dependency can be shortened.06-04-2009
20090012920Human Artificial Intelligence Software Program - A method of creating human artificial intelligence in machines and computer software is presented here, as well as methods to simulate human reasoning, thought and behavior. The present invention serves as a universal artificial intelligence program that will store, retrieve, analyze, assimilate, predict the future and modify information in a manner and fashion which is similar to human beings and which will provide users with a software application that will serve as the main intelligence of one or a multitude of computer based programs, software applications, machines or compilation of machinery.01-08-2009
20090012919EXPLAINING CHANGES IN MEASURES THRU DATA MINING - Systems and methodologies for identification of factors that cause significant shifts in transactions in a relational store and/or OLAP environment. Transactions are grouped into significant categories defined across the whole data space, to detect interesting sub spaces transactions. Subsequently, sub spaces that show strong variance between two slices can be selected, followed by grouping the subspaces in sub reports to measure the coverage for each sub report. A final report can then be generated that contains list of sub-reports detected in the previous acts.01-08-2009
20110225107SEMANTICS UPDATE AND ADAPTIVE INTERFACES IN CONNECTION WITH INFORMATION AS A SERVICE - Additional semantic information that describes data sets is inferred in response to a request for data from the data sets, e.g., in response to a query over the data sets, including analyzing a subset of results extracted based on the request for data to determine the additional semantic information. The additional semantic information can be verified by the publisher as correct, or satisfy correctness probabilistically. Mapping information based on the additional semantic information can be maintained and updated as the system learns additional semantic information (e.g., information about what a given column represents and data types represented), and the form of future data requests (e.g., URL based queries) can be updated to more closely correspond to the updated additional semantic information.09-15-2011
20090204556Large Scale Manifold Transduction - A method for training a learning machine for use in discriminative classification and regression includes randomly selecting, in a first computer process, an unclassified datapoint associated with a phenomenon of interest; determining, in a second computer process, a set of datapoints associated with the phenomenon of interest that is likely to be in the same class as the selected unclassified datapoint; predicting, in a third computer process, a class label for the selected unclassified datapoint in a third computer process; predicting a class label for the set of datapoints in a fourth computer process; combining the predicted class labels in a fifth computer process, to predict a composite class label that describes the selected unclassified datapoint and the set of datapoints; and using the combined class label to adjust at least one parameter of the learning machine in a sixth computer process.08-13-2009
20090198635OPTICAL METROLOGY OF STRUCTURES FORMED ON SEMICONDUCTOR WAFERS USING MACHINE LEARNING SYSTEMS - A structure formed on a semiconductor wafer is examined by obtaining a first diffraction signal measured using a metrology device. A second diffraction signal is generated using a machine learning system, where the machine learning system receives as an input one or more parameters that characterize a profile of the structure to generate the second diffraction signal. The first and second diffraction signals are compared. When the first and second diffraction signals match within a matching criterion, a feature of the structure is determined based on the one or more parameters or the profile used by the machine learning system to generate the second diffraction signal.08-06-2009
20090083199PROCESSING DEVICE HAVING SELECTIBLE LIST ITEMS WITH INTUITIVE LEARNING CAPABILITY - A processing device and a method of providing learning capability thereto are provided. A list containing a plurality of listed items with an associated item probability distribution is generated. The item probability distribution comprises a plurality of probability values corresponding to the plurality of listed items. One or more items are selected from the plurality of listed items based on the item probability distribution, a performance index indicative of a performance of the processing device relative to the objective is determined, and the item probability distribution is modified based on the performance index.03-26-2009
20090063374Category Classification Method - A category classification method includes: calculating function values corresponding to a relationship between a classification target and support vectors that contribute to a classification boundary, calculating an addition value in which the function value for each support vector has been added, and classifying that the classification target does not pertain to a specific category in case that the addition value is smaller than a threshold, wherein calculation of the addition value is carried out by adding function values having positive values, then adding function values having negative values, and the classification target is classified as not pertaining to the specific category, without adding the remaining function values, in case that the addition value has become smaller than the threshold.03-05-2009
20090248596CONFIGURATION INFORMATION MANAGEMENT APPARATUS, CONFIGURATION INFORMATION MANAGEMENT PROGRAM, AND CONFIGURATION INFORMATION MANAGEMENT METHOD - A CMDB (configuration information management database) stores a CI (configuration item) and know-how separately. A CMDB data update management unit associates with each set of “property:value” stored in the CI with related know-how in the CMDB. The know-how stores a set of “property:value” common to a number of associated CIs.10-01-2009
20110145178DATA CLASSIFICATION USING MACHINE LEARNING TECHNIQUES - A system and article of manufacture enabling adapting to a shift in document content according to one embodiment of the present invention includes instructions for: receiving at least one labeled seed document; receiving unlabeled documents; receiving at least one predetermined cost factor; training a transductive classifier using the at least one predetermined cost factor, the at least one seed document, and the unlabeled documents; classifying the unlabeled documents having a confidence level above a predefined threshold into a plurality of categories using the classifier; reclassifying at least some of the categorized documents into the categories using the classifier; and outputting identifiers of the categorized documents to at least one of a user, another system, and another process. Systems and articles of manufacture for separating documents are also presented. Systems and articles of manufacture for document searching are also presented.06-16-2011
20110145175Methods, Systems and Media Utilizing Ranking Techniques in Machine Learning - Methods, systems and media are taught utilizing ranking techniques in machine learning to learn a ranking function. Specifically, ranking algorithms are applied to learn a ranking function that advantageously minimizes ranking error as a function of targeted ranking order discrepancies between a predetermined first ranking of a training plurality of data elements and a second ranking of the training plurality of data elements by the ranking function. The ranking algorithms taught may be applied to ranking representations of chemical structures and may be particularly advantageous in the field of drug discovery, e.g., for prioritizing chemical structures for drug screenings.06-16-2011
20090063375SYSTEM AND METHOD FOR COMPILING RULES CREATED BY MACHINE LEARNING PROGRAM - A system, a method, and a machine-readable medium are provided. A group of linear rules and associated weights are provided as a result of machine learning. Each one of the group of linear rules is partitioned into a respective one of a group of types of rules. A respective transducer for each of the linear rules is compiled. A combined finite state transducer is created from a union of the respective transducers compiled from the linear rules.03-05-2009
20110231351Feedback in Group Based Hierarchical Temporal Memory System - A Hierarchical Temporal Memory (HTM) network has at least first nodes and a second node at a higher level than the first nodes. The second node provides an inter-node feedback signal to the first nodes for grouping patterns and sequences (or co-occurrences) in input data received at the first nodes at the first nodes. The second node collects forward signals from the first nodes; and thus, the second node has information about the grouping of the patterns and sequences (or co-occurrences) at the first nodes. The second node provides inter-node feedback signals to the first nodes based on which the first nodes may perform the grouping of the patterns and sequences (or co-occurrences) at the first nodes.09-22-2011
20100153317Intelligent robot and control method thereof - Disclosed herein are a robot with a judgment system to enable implementation of multi-dimensional recognitions, thoughts and actions, and a control method thereof. The judgment system includes a dialog system and a task-planning system. The dialog system includes a dialog manager to manage the progress of a dialog of the intelligent robot with a user. The task-planning system includes a leader agent, an action agent and an interaction agent and serve to control a goal, plan and action of a task to be performed by the intelligent robot based on the dialog. The judgment system assists separation of concerns and consequently, enhances convenience of development. The judgment system of the robot contains a mechanism that considers a great number of cases, such as a task priority, immediate user input, inherent robot task, etc., enabling implementation of multi-dimensional recognitions, thoughts and actions.06-17-2010
20120143797Metric-Label Co-Learning - Labels for unlabeled media samples may be determined automatically. Characteristics and/or features of an unlabeled media sample are detected and used to iteratively optimize a distance metric and one or more labels for the unlabeled media sample according to an algorithm. The labels may be used to produce training data for a machine learning process.06-07-2012
20120143790RELEVANCE OF SEARCH RESULTS DETERMINED FROM USER CLICKS AND POST-CLICK USER BEHAVIOR OBTAINED FROM CLICK LOGS - Data from a click log may be used to generate training data for a search engine. User click behavior and user post-click behavior may be used to assess the relevance of a page to a query. Labels for training data may be generated based on data from the click log. The labels may pertain to the relevance of a page to a query. For example, user post-click behavior that may be examined includes the amount of time that a user remains on a target page when a user clicks one of the search results.06-07-2012
20120078828GAS BLOCKING DEVICE - A newly-purchased gas appliance is detected and reported to a gas administrator. There are provided a flow rate detection portion, a flow rate calculation portion, a code extraction portion, an initial code learning portion, a code maintaining portion, a code judging portion, an additional code learning portion, and an external communication portion. The code extraction portion extracts a code pattern E. The initial code learning portion gathers similar code patterns E as a gas appliance code pattern F. The code judging portion judges whether or not the code pattern E matches any of gas appliance code patterns F held by the code maintaining portion within a predetermined range. The code patterns E that have failed to match are subjected to additional identification of a gas appliance in the additional code learning portion. The gas blocking device can thereby let the additional code learning portion detect whether or not a new gas appliance has emerged and the external communication portion send a report to the gas administrator.03-29-2012
20120078827HIERARCHICAL TEMPORAL MEMORY METHODS AND SYSTEMS - Methods and systems for constructing biological-scale hierarchically structured cortical statistical memory systems utilizing fabrication technology and meta-stable switching devices. Learning content-addressable memory and statistical random access memory circuits are detailed. Additionally, local and global signal modulation of bottom-up and top-down processing for the initiation and direction of behavior is disclosed.03-29-2012
20120078826FACT CHECKING USING AND AIDING PROBABILISTIC QUESTION ANSWERING - A system, a method and a computer program product for verifying a statement are provided. The system is configured to receive a statement. The system is configured to decompose the received statement into one or more sets of question and answer pairs. The system is configured to determine a confidence value of each answer in the one or more question and answer pair sets. The system is configured to combine the determined confidence values. The combined confidence values represent a probability that the received statement is evaluated as true.03-29-2012
20120078825SEARCH RESULT RANKING USING MACHINE LEARNING - Various embodiments include systems and methods for search result ranking using machine learning. A goal model can be created using machine learning. Responsive to a search query, a plurality of data factors can be inputted into the goal model to create a model output. Search results can be presented to a user based on the model output.03-29-2012
20110231347Named Entity Recognition in Query - Named Entity Recognition in Query (NERQ) involves detection of a named entity in a given query and classification of the named entity into one or more predefined classes. The predefined classes may be based on a predefined taxonomy. A probabilistic approach may be taken to detecting and classifying named entities in queries, the approach using either query log data or click through data and Weakly Supervised Latent Dirichlet Allocation (WS-LDA) to construct and train a topic model.09-22-2011
20120078824Method and System for Music Recommendation Based on Immunology - An artificial intelligence song/music recommendation system and method is provided that allows music shoppers to discover new music. The system and method accomplish these tasks by analyzing a database of music in order to identify key similarities between different pieces of music, and then recommends pieces of music to a user depending upon their music preferences. Once the song files have been analyzed and mapped, this system uses four layers, metaphorically equivalent to the human immune system, to provide music recommendation.03-29-2012
20120078823ABNORMALITY DIAGNOSIS FILTER GENERATOR - Provided is an apparatus determining values of N and K for an abnormality diagnostic logic which makes a diagnosis N times for each diagnosis target by using observation values collected therefrom, and generates a diagnosis result showing that the diagnosis target is abnormal if the diagnosis target is judged to be abnormal K or more times. A calculator calculates average false detection rate P03-29-2012
20120078821METHODS FOR UNSUPERVISED LEARNING USING OPTIONAL POLYA TREE AND BAYESIAN INFERENCE - The present disclosure describes an extension of the Pólya Tree approach for constructing distributions on the space of probability measures. By using optional stopping and optional choice of splitting variables, the present invention gives rise to random measures that are absolutely continuous with piecewise smooth densities on partitions that can adapt to fit the data. The resulting optional Pólya tree distribution has large support in total variation topology, and yields posterior distributions that are also optional Pólya trees with computable parameter values.03-29-2012
20100121791SYSTEM, METHOD AND PROGRAM FOR PHARMACOKINETIC PARAMETER PREDICTION OF PEPTIDE SEQUENCE BY MATHEMATICAL MODEL - The present invention relates to the system, method and program for the pharmacokinetic parameter prediction of peptide sequence by the mathematical model.05-13-2010
20100121790METHOD, APPARATUS AND COMPUTER PROGRAM PRODUCT FOR CATEGORIZING WEB CONTENT - An apparatus for providing web content categorization may include a processor configured to receive an indication of a web page to be evaluated, evaluate the web page based on characteristics of the web page in relation to previously categorized web pages assigned to respective ones of a structured group of categories, and assign the web page to at least one of the categories based on the evaluation. A corresponding method and computer program product are also provided.05-13-2010
20110231350ACTIVE METRIC LEARNING DEVICE, ACTIVE METRIC LEARNING METHOD, AND ACTIVE METRIC LEARNING PROGRAM - An active metric learning device includes a metric application data analysis unit, a metric optimization unit, and an attribute clustering unit. The metric application data analysis unit is formed with a metric applying module for calculating the distance between data to be analyzed, a data analyzing module for analyzing the data using a predetermined function and the distances between the data to be analyzed and outputting the result of the data analysis, and an analysis result storage unit for storing the result of the data analysis. The metric optimization unit is formed with a feedback converting module for creating side information according to the command of feedback from the user and a metric learning module for generating a metric matrix optimized under a predetermined condition using the created side information. The attribute clustering unit clusters the metric matrix optimized by the metric optimization unit and structuralizes the attributes.09-22-2011
20110231349SYSTEMS AND METHODS OF COGNITIVE PATTERNS KNOWLEDGE GENERATION - A processor based system and method of generating cognitive pattern knowledge of a sensory input is disclosed. The method comprising the steps of receiving sensory input to create at least one concrete pattern, receiving at least one abstract pattern comprising abstract segments and vertically blending the concrete pattern with the abstract pattern by selectively projecting abstract segments to create a vertically blended pattern whereby the vertically blended pattern represents cognitive pattern knowledge of the sensory input. In some embodiments, the systems and methods further comprise creating a measure of a degree of vertical blending and when the measure of the degree of vertical blending exceeds a threshold, horizontally blending at least two abstract patterns to create a horizontally blended abstract pattern.09-22-2011
20110231348Regularized Dual Averaging Method for Stochastic and Online Learning - Described is a technology by which a learned mechanism is developed by solving a minimization problem by using regularized dual averaging methods to provide regularized stochastic learning and online optimization. An objective function sums a loss function of the learning task and a regularization term. The regularized dual averaging methods exploit the regularization structure in an online learning environment, in a manner that obtains desired regularization effects, e.g., sparsity under L1-regularization.09-22-2011
20080262987OBTAINING A VALUE VIA A RULE ENGINE IMPLEMENTED BY A COLLECTION OBJECT - A system and computer program product for obtaining a value via a rule engine implemented by a collection object associated with an object-oriented application. A request for the value includes a key, is received from the application, and is directed to a method of the collection object. The collection object is capable of storing the key and associated data, and providing the data in response to receiving the request. An overriding of the method of the collection object replaces the provision of the data by the collection object with a processing of the request by a rule engine external to the application. The rule is identified in a rule definition file external to the application based on an association between the rule and the key. An algorithm associated with the rule and included in the rule definition file is executed to provide the requested value.10-23-2008
20080262986METHOD FOR TRAINING A CLASSIFIER - A method for training a classifier which forms part of a search engine comprises: receiving a document submitted by an end user of the search engine at a server; creating a training set of documents, the training set including the document submitted by the end user; training the classifier using the training set; and paying an incentive to the end user for submitting the document.10-23-2008
20080262985SYSTEMS, METHODS, AND MEDIA FOR GENERATING SANITIZED DATA, SANITIZING ANOMALY DETECTION MODELS, AND/OR GENERATING SANITIZED ANOMALY DETECTION MODELS - Systems, methods, and media for generating sanitized data, sanitizing anomaly detection models, and generating anomaly detection models are provided. In some embodiments, methods for generating sanitized data are provided. The methods including: dividing a first training dataset comprised of a plurality of training data items into a plurality of data subsets each including at least one training data item of the plurality of training data items of the first training dataset; based on the plurality of data subsets, generating a plurality of distinct anomaly detection micro-models; testing at least one data item of the plurality of data items of a second training dataset of training data items against each of the plurality of micro-models to produce a score for the at least one tested data item; and generating at least one output dataset based on the score for the at least one tested data item.10-23-2008
20090210364Apparatus for and Method of Generating Complex Event Processing System Rules - A novel and useful mechanism enabling a standard learning algorithm to generate rules for complex event processing (CEP) systems. The method creates rules that infer previously defined output events by creating input event feature vectors for each targeted output event. In addition, a method for automatically generating CEP system rules to infer output events which are anomalies (i.e. statistical outliers) of input event sequences is disclosed. Input feature vectors consisting of multiple input events and parameters for each targeted output event are then input into a standard learning algorithm to generate CEP system rules.08-20-2009
20090210363SYSTEMS, METHODS AND COMPUTER PROGRAM PRODUCTS FOR SUPERVISED DIMENSIONALITY REDUCTION WITH MIXED-TYPE FEATURES AND LABELS - Systems, methods and computer program products for supervised dimensionality reduction. Exemplary embodiments include a method including receiving an input in the form of a data matrix X of size N×D, wherein N is a number of samples, D is a dimensionality, a vector Y of size N×1, hidden variables U of a number K, a data type of the matrix X and the vector Y, and a trade-off constant alpha; selecting loss functions in the form of Lx(X,UV) and Ly(Y,UW) appropriate for the type of data in the matrix X and the vector Y, where U, V and W are matrices, selecting corresponding sets of update rules RU, RV and RW for updating the matrices U,V and W, learning U, V and W that provide a minimum total loss L(U,V,W)=Lx(X,UV)+alpha*Ly(Y,UW), and returning matrices U, V and W.08-20-2009
20110125679METHOD FOR APPROXIMATING USER TASK REPRESENTATIONS BY DOCUMENT-USAGE CLUSTERING - Embodiments of the present invention provide a system for automatically creating a task representation associated with a user task. The system calculates usage footprints of a document based on other applications, documents, and people that have been accessed by the user within a predetermined time frame before and after the user accesses the document. After obtaining usage footprints of a number of documents, the system applies a clustering technique, such as spectral clustering, to create task representations, each including a collection (cluster) of documents and/or applications that are used for accomplishing a particular task. The system also filters the documents based on their average dwell times, and uses user feedback to merge or split different task clusters in order to provide accurate task representations.05-26-2011
20090106175MANAGEMENT OF APPLICATIVE STREAMS IN MOBILE NETWORKS - A method is provided for constructing at least one decision graph for managing at least one applicative stream assigned to a terminal and set up between the terminal and a correspondent via at least one communication network. The method includes a step of dynamically constructing at least one possible decision graph for the one applicative stream assigned to the terminal, itself including a step of exchanging at least one configuration message between at least two decision modules pertaining to a predetermined set of decision modules.04-23-2009
20090106172FALSE DISCOVERY RATE FOR GRAPHICAL MODLES - The claimed subject matter provides systems and/or methods that determines a number of non-spurious arcs associated with a learned graphical model. The system can include devices and mechanisms that utilize learning algorithms and datasets to generate learned graphical models and graphical models associated with null permutations of the datasets, ascertaining the average number of arcs associated with the graphical models associated with null permutations of the datasets, enumerating the total number of arcs affiliated with the learned graphical model, and presenting a ratio of the average number of arcs to the total number of arcs, the ratio indicative of the number of non-spurious arcs associated the learned graphical model.04-23-2009
20090204555SYSTEM AND METHOD USING HIDDEN INFORMATION - A method and system for use in describing a phenomenon of interest. The method and system computes a decision rule for use in describing the phenomenon of interest using training data relating to the phenomenon of interest, labels for labeling the training data, and hidden information about the training data or directed distances obtained from the hidden information, as inputs.08-13-2009
20090204554Direction-aware proximity for graph mining - A method and system for graph mining direction-aware proximity measurements. A directed graph includes nodes and directed edges connecting the nodes. A direction-aware proximity measurement is calculated from a first node to a second node or from a first group of nodes to a second group of nodes. The direction-aware proximity measurement from a first node to second node is based on an escape probability from the first node to the second node. Disclosed herein are methods for efficiently calculating one or multiple direction-aware proximity measurements. The direction-aware proximity measurements can be used in performing various graph mining applications.08-13-2009
20090204553Feature Reduction Method for Decision Machines - A method for feature reduction in a training set for a learning machine such as a Support Vector Machine (SVM). In one embodiment the method includes a step (08-13-2009
20090210362OBJECT DETECTOR TRAINED USING A WORKING SET OF TRAINING DATA - An object detector that includes a number of weak classifiers can be trained using a subset (a “working set”) of training data instead of all of the training data. The working set can be updated so that, for example, it remains representative of the training data. A decision to update the working set may be made based on the false positive sample rate—if that rate falls below a threshold value, an update of the working set can be triggered.08-20-2009
20090240640Apparatus and method for predicting engine test performance from bench test data - An apparatus for evaluating deposit formation characteristics of lubricant samples. The apparatus includes a reactor chamber having a closed first end and a second end, said closed first end of said reactor chamber forming a lubricant sump and said second end open to the atmosphere, an electric heater for heating a test coupon positioned thereon, said electric heater positioned within said reactor chamber, above said lubricant sump, an air supply tube, said air supply tube having a first end in fluid communication with a source of air and a second end, said second end positioned adjacent the test coupon, said air supply tube having a sample orifice located between said first end and said second end of said air supply tube and a lubricant supply tube, said lubricant supply tube having a first end positioned within said lubricant sump and in fluid communication with a source of lubricant and a second end, said second end positioned within said sample orifice of said air supply tube. A method of predicting lubricant deposit formation in an end use engine test and a method for predicting whether a candidate lubricant sample will pass an End Use Test are also provided.09-24-2009
20090222388Method of and system for hierarchical human/crowd behavior detection - The present invention is directed to a computer automated method of selectively identifying a user specified behavior of a crowd. The method comprises receiving video data but can also include audio data and sensor data. The video data contains images a crowd. The video data is processed to extract hierarchical human and crowd features. The detected crowd features are processed to detect a selectable crowd behavior. The selected crowd behavior detected is specified by a configurable behavior rule. Human detection is provided by a hybrid human detector algorithm which can include Adaboost or convolutional neural network. Crowd features are detected using textual analysis techniques. The configurable crowd behavior for detection can be defined by crowd behavioral language.09-03-2009
20090222387Diagnosis, Prognosis and Prediction of Recurrence of Breat Cancer - The present invention relates to methods and compositions for the diagnosis, prognosis, and prediction of breast cancer. More specifically, the invention relates to classification of breast cancer tissue samples based on measuring the expression of a set of marker genes. The set is useful for the identification of clinically important breast cancer subtypes. Methods are disclosed for prediction, diagnosis and prognosis of breast cancer.09-03-2009
20090222389CHANGE ANALYSIS SYSTEM, METHOD AND PROGRAM - Different virtual labels, for example, like +1 and −1, are assigned to two data sets. A change analysis problem for the two data sets is reduced to a supervised learning problem by using the virtual labels. Specifically, a classifier such as logical regression, decision tree and SVM is prepared and is trained by use of a data set obtained by merging the two data sets assigned the virtual labels. A feature selection function of the resultant classifier is used to rank and output both every attribute contributing to classification and its contribution rate.09-03-2009
20090240638SYNTACTIC AND/OR SEMANTIC ANALYSIS OF UNIFORM RESOURCE IDENTIFIERS - Subject matter disclosed herein may relate to analyses of uniform resource identifiers associated with web pages, and further may relate to gathering information about web pages by analyzing the uniform resource identifiers.09-24-2009
20100274745PREDICTION METHOD FOR MONITORING PERFORMANCE OF POWER PLANT INSTRUMENTS - Disclosed is a prediction method for monitoring performance of power plant instruments. The prediction method extracts a principal component of an instrument signal, obtains an optimized constant of a SVR model through a response surface methodology using data for optimization, and trains a model using training data. Therefore, compared to an existing Kernel regression method, accuracy for calculating a prediction value can be improved.10-28-2010
20090254498System and method for identifying critical emails - Disclosed is a method and system for identifying critical emails. To identify critical emails, a critical email classifier is trained from training data comprising labeled emails. The classifier extracts N-grams from the training data and identifies N-gram features from the extracted N-grams. The classifier also extracts salient features from the training data. The classifier is trained based on the identified N-gram features and the salient features so that the classifier can classify unlabeled emails as critical emails or non-critical emails.10-08-2009
20090254501WORD-SPACING CORRECTION SYSTEM AND METHOD - A word-spacing correction system and method are provided to automatically recognize and correct errors in the spacing of word inputs in an electronic device with relatively low computing power. In a learning process, probability information about each feature is created from a corpus of correct words, and then error correction rules are created by applying the probability information to a corpus of incorrect words from which all spaces between words of the corpus of correct words are removed. In an applying process, word-spacing in a user's input sentence is corrected by applying the probability information and the error correction rules to the user's input sentence.10-08-2009
20090254499TECHNIQUES TO FILTER MEDIA CONTENT BASED ON ENTITY REPUTATION - Techniques to filter media content based on entity reputation are described. An apparatus may comprise a reputation subsystem operative to manage an entity reputation score for an entity. The reputation subsystem comprising a reputation manager component and a reputation input/output (I/O) component. The reputation manager component may comprise, among other elements, a data collection module operative to collect reputation information for an entity from a selected set of multiple reputation sources. The reputation manager component may also comprise a feature manager module communicatively coupled to the data collection module, the feature manager module operative to extract a selected set of reputation features from the reputation information. The reputation manager component may further comprise a reputation scoring module communicatively coupled to the feature manager module, the reputation scoring module operative to generate an entity reputation score based on the reputation features using a supervised or unsupervised machine learning algorithm. Other embodiments are described and claimed.10-08-2009
20100153319METHODS, APPARATUS AND ARTICLES OF MANUFACTURE TO CHARACTERIZE APPLICATIONS - Example methods, apparatus and articles of manufacture to characterize applications are disclosed. A disclosed example method includes collecting resource utilization trace data from the two or more applications simultaneously running on one or more computational devices, determining an intrinsic dimensionality of the collected trace data, the intrinsic dimensionality representing a number of predominate features that substantially characterize the trace data, and characterizing each application's workload based on the determined intrinsic dimensionality.06-17-2010
20100153314SYSTEMS AND METHODS FOR COLLABORATIVE FILTERING USING COLLABORATIVE INDUCTIVE TRANSFER - Systems and methods are disclosed that are configured to access a database that includes a list of members of a first group, a list of members of a second group, and ratings for at least some of the members of the second group. The ratings are attributed to the members of the first group. A machine learning training set is built for a particular member of the first group. The training set includes class labels corresponding to the particular member's ratings for the members of the second group, and features that include supplied and predicted ratings from at least a subset of processed members of the first group. A predictor for the particular member of the first group is trained based on the machine learning training set. The predictor corresponding to the particular member is used to generate predicted ratings for one or more members of the second group the particular member has not rated.06-17-2010
20090276383RULES GENERATION FOR IT RESOURCE EVENT SITUATION CLASSIFICATION - A computer processing device receives computer readable data to derive computer executable rules for mining and constructing situation categories. The received data is transformed into a predetermined standard format if the received data is not already in the predetermined standard format. The predetermined standard formatted data is parsed, and an outer, iterative loop is performed until at least one predetermined stopping criterion is met. An inner iterative loop is performed within the outer iterative loop until all desired subsets of data are processed. During the inner iterative loop, selected subsets of data are labeled with labels associated with corresponding previously labeled subsets of data. New computer executable rules are generated for mining and constructing situation categories from the labeled subsets of data. Keyword list classifiers are transformed using the stored labeled subsets of data.11-05-2009
20100153320METHOD AND ARRANGEMENT FOR SIM ALGORITHM AUTOMATIC CHARSET DETECTION - The invention relates, in an embodiment, to a computer-implemented method for handling a target document, the target document having been transmitted electronically and involving an encoding scheme. The method includes training, using a plurality of text document samples, to obtain a set of machine learning models. Training includes using SIM (Similarity Algorithm) to generate the set of machine learning models from feature vectors obtained from the plurality of text document samples. The method also includes applying the set of machine learning models against a set of target document feature vectors converted from the target document to detect the encoding scheme. The method including decoding the target document to obtain decoded content of the document based on at least the first encoding scheme.06-17-2010
20100153316SYSTEMS AND METHODS FOR RULE-BASED ANOMALY DETECTION ON IP NETWORK FLOW - A system to detect anomalies in internet protocol (IP) flows uses a set of machine-learning (ML) rules that can be applied in real time at the IP flow level. A communication network has a large number of routers that can be equipped with flow monitoring capability. A flow collector collects flow data from the routers throughout the communication network and provides them to a flow classifier. At the same time, a limited number of locations in the network monitor data packets and generate alerts based on packet data properties. The packet alerts and the flow data are provided to a machine learning system that detects correlations between the packet-based alerts and the flow data to thereby generate a series of flow-level alerts. These rules are provided to the flow time classifier. Over time, the new packet alerts and flow data are used to provide updated rules generated by the machine learning system.06-17-2010
20080306890Plant Control Apparatus - A plant control system includes: a numerical calculation execution part which calculates the operation characteristic of the plant; a model for simulating the plant control characteristic according to information on the numerical calculation result; a learning part which learns the plant operation method by using the model; a learning information database which stores learning information data on the learning part; a pattern generation part which generates pattern data expressing a state input based on the learning information data in the learning part with a smaller input number than the model input dimension; a pattern database which stores the pattern data generated in the pattern generation part; and a learning result determination part which selects a learning result having a preferable control effect from the learning result obtained by using a plurality of patterns.12-11-2008
20080306891METHOD FOR MACHINE LEARNING WITH STATE INFORMATION - Methods, systems, and computer program products are provided for the online convex optimization problem, in which the decision maker has knowledge of the all past states and resulting cost functions for his previous choices and attempts to make a new choice that results in minimum regret. The method does not rely upon the structure of the cost function or the characterization of the states and takes advantage of the similarity between successive states to enable the method to converge to a reasonably optimal result.12-11-2008
20080306889SYSTEM FOR SUPPORTING USER'S BEHAVIOR - Provided is a system 12-11-2008
20080306888STOCHASTIC CONTROL OPTIMIZATION FOR SENDER-BASED FLOW CONTROL IN A DISTRIBUTED STATEFUL MESSAGING SYSTEM - A method and system for controlling message flow in distributed stream processing. State transition probabilities in a Markov model having one state per staleness value of data are determined for sending or withholding updates of data to subscribers using expected message rates from an information provider. A cost function annotates each state transition in the model with a state transition cost for each decision to “send” or “withhold”. A propagation policy specifying whether to send or withhold the message is determined for each state. The propagation policy is then deployed. If a new message comprising an update of data is received during a lapsed time unit, a staleness value of the data held by subscribers is increased. The propagation policy is used to determine whether to send or withhold the message. If the message should be sent, the message is propagated and the staleness value of the data is reset.12-11-2008
20080306887METHOD FOR MACHINE LEARNING WITH STATE INFORMATION - Methods, systems, and computer program products are provided for the online convex optimization problem, in which the decision maker has knowledge of the all past states and resulting cost functions for his previous choices and attempts to make a new choice that results in minimum regret. The method does not rely upon the structure of the cost function or the characterization of the states and takes advantage of the similarity between successive states to enable the method to converge to a reasonably optimal result.12-11-2008
20090094174METHOD, SYSTEM AND PROGRAM PRODUCT FOR ON DEMAND DATA MINING SERVER WITH DYNAMIC MINING MODELS - The present invention in various implementations provides a method, system and computer program product for dynamically determining data mining results using a dynamic data mining model within a data mining system. The present invention, in accordance with various implementations, in part, creates a mining model for an event request that includes a plurality of mining rule sets determined in relation to the event and one or more business objectives and selected computations.04-09-2009
20090089225WEB-BASED VISUALIZATION MASH-UPS FOR INDUSTRIAL AUTOMATION - A visualization system that generates visual mash-ups for industrial automation includes a ash-up component that combines output from a subset of disparate sources into a common interface. The disparate sources include at least one of equipment, computers, or devices within an industrial automation environment. A visualization component generates and displays a mash-up visualization that includes information associated with the common interface.04-02-2009
20100179931DEVELOPING SYSTEM THINKERS - The system thinker application receives a first issue, a first resolution to the first issue, and a first plurality of skills. The system thinker application searches a system environment electronic profile for a second issue, a second resolution to the second issue, and a second plurality of skills, wherein the system environment electronic profile contains a plurality of component profiles, and wherein the plurality of component profiles contain a second issue, a second resolution to the second issue, and a second plurality of skills. The system thinker application determines if the first issue, the first resolution to the first issue, and any one of the first plurality of skills are similar to any one of the second issue, the second resolution to the second issue, and any one of the second plurality of skills. The system thinker application adds skills to the system environment electronic profile and the component profile.07-15-2010
20120123979PERSON EVALUATION DEVICE, PERSON EVALUATION METHOD, AND PERSON EVALUATION PROGRAM - A person evaluation device includes a collecting unit that collects event data in which activities performed by members are recorded; a creating unit that creates a combination of evaluation programs each of which calculates an evaluation value of a person to be evaluated in accordance with a value that is set in a predetermined item contained in evaluation items contained in the event data, a calculating unit that calculates a coverage percentage that represents a percentage that is used to calculate the evaluation value by at least one evaluation program in which event data, from among the collected event data, related to a member associated with the person to be evaluated is included in the created combination, and an output unit that outputs information related to the evaluation program included in the combination that is selected in accordance with the calculated coverage percentage.05-17-2012
20120123976Object-Sensitive Image Search - Methods and systems for object-sensitive image searches are described herein. These methods and systems are usable for receiving a query for an image of an object and providing a ranked list of query results to the user based on a ranking of the images. The object-sensitive image searches may generate a pre-trained multi-instance learning (MIL) model trained from free training data from users sharing images at websites to identify a common pattern of the object, and/or may generate a MIL model “on the fly” trained from pseudo-positive and pseudo-negative samples of query results to identify a common pattern of the object. As such, the user is presented with query results that include images that prominently display the object near the top of the results.05-17-2012
20100179933SUPERVISED SEMANTIC INDEXING AND ITS EXTENSIONS - A system and method for determining a similarity between a document and a query includes building a weight vector for each of a plurality of documents in a corpus of documents stored in memory and building a weight vector for a query input into a document retrieval system. A weight matrix is generated which distinguishes between relevant documents and lower ranked documents by comparing document/query tuples using a gradient step approach. A similarity score is determined between weight vectors of the query and documents in a corpus by determining a product of a document weight vector, a query weight vector and the weight matrix.07-15-2010
20100161528Method Of and Apparatus For Automated Behavior Prediction - A computer-implemented method of behavior prediction includes selecting behavior examples having corresponding antecedent candidates, identifying source text descriptions describing the behavior examples, automatically extracting predictors as common themes across all statements and all behavior examples with a language-independent theme extraction process, flagging each behavior example to indicate a presence or absence of the corresponding extracted antecedents in each of the source text descriptions and creating a data array consisting of antecedent columns and behavior example rows, submitting the data array to a pattern classifier to extract patterns among the antecedent candidates and outcomes by training and validating the pattern classifier and predicting a new occurrence of a target behavior by entering a current state of the antecedents to the trained pattern classifier.06-24-2010
20100262570Information Processing Apparatus and Method, and Program Thereof - There is provided an information processing apparatus including: evaluation information extracting means extracting evaluation information from evaluation of every user for an item; preference information creating means for creating preference information indicating a preference of every user on the basis of the evaluation information extracted by the evaluation information extracting means and an item characteristic amount indicating a characteristic of the item; space creating means for creating a space in which the user is located, according to the preference information; and display control means for controlling display of the user located in the space, according to the space created by the space creating means and the preference information. The apparatus may be applied to, for example, an image display apparatus which displays server images for providing a variety of items and information.10-14-2010
20100185570THREE-DIMENSIONAL MOTION IDENTIFYING METHOD AND SYSTEM - A three-dimensional (3D) motion identifying method and system are used for identifying a motion of an object in a 3D space. First, the method provides a database recording sets of predetermined inertial information, and each of the sets of predetermined inertial information is an inertial movement of a specific motion in the 3D space. Then, inertial information of the object in moving is retrieved via a motion sensor in the object, and the inertial information is compared with all predetermined inertial information in the database to determine similarities therebetween. Finally, whether the motion of the object is the same as any predetermined inertial information or not is determined according to a degree of the similarity. As a result, more complicated motions of the object can be directly identified via the comparison with the database.07-22-2010
20100185569Smart Attribute Classification (SAC) for Online Reviews - Techniques for identifying attributes in a sentence and determining a number of attributes to be associated with the sentence is described.07-22-2010
20080215510MULTI-SCALE SEGMENTATION AND PARTIAL MATCHING 3D MODELS - A scale-Space feature extraction technique is based on recursive decomposition of polyhedral surfaces into surface patches. The experimental results show that this technique can be used to perform matching based on local model structure. Scale-space techniques can be parameterized to generate decompositions that correspond to manufacturing, assembly or surface features relevant to mechanical design. One application of these techniques is to support matching and content-based retrieval of solid models. Scale-space technique can extract features that are invariant with respect to the global structure of the model as well as small perturbations that 09-04-2008
20080215511SYSTEM AND METHOD FOR AUTOMATED PART-NUMBER MAPPING - Automated mapping of part numbers associated with parts in a bill of materials (BOM) submitted by a BOM originator to internal part numbers assigned to those parts by a BOM receiver is performed by one or more computers connected to one or more networks through one or more network interfaces. A first receive component receives one or more data sets containing historical data on bills of materials received in the past by the BOM receiver. A second receive component receives one or more data sets containing known mappings between internal part numbers used by the BOM receiver, and part numbers used by various BOM originators. A third receive component receives one or more data sets containing information of various parameters and their values describing the parts to which the BOM receiver has assigned internal part numbers. A fourth receive component receives one or more methods of automatically learning models for predicting internal part numbers from the above mentioned historical BOM data, mapping data and part parametric data. A learning component learns the models from the data. A fifth receive component receives a BOM from a requesting process. The BOM has one or more parts with a missing internal part number. A mapping component applies the learned models to the received BOM to automatically determine internal part numbers for all unmapped BOM originator part numbers. A release process assigns internal part numbers to all unmapped parts in the BOM and releases the BOM to the requesting process.09-04-2008
20100191681Prognostics and health management method for aging systems - The present invention provides a novel prognostic and health management method for natural aging systems. This prognostic and health management method can detect anomalies in a system in advance, and can determine whether the detected anomalies are due to natural aging or other aging processes. In this prognostic method, a moving window method for improving the performance of the conventional data-driven prognostic methods is described. This prognostic and health management method combines with the detections by the data-driven prognostic method based on the conventional training and moving window methods to determine whether the detected anomalies are due to natural aging or other aging processes and in so doing can reduce the number of false alarms; reduce cost of a system by decreasing the unnecessary maintenance, downtime, and inventory; can extend the life of systems; and can assist in the design and qualification of future systems to improve their reliability.07-29-2010
20100179934 KERNEL-BASED METHOD AND APPARATUS FOR CLASSIFYING MATERIALS OR CHEMICALS AND FOR QUANTIFYING THE PROPERTIES OF MATERIALS OR CHEMICALS IN MIXTURES USING SPECTROSCOPIC DATA - A kernel-based method determines the similarity of a first spectrum and a second spectrum. Each spectrum represents a result of spectral analysis of a material or chemical and comprises a set of spectral attributes distributed across a spectral range. The method calculates a kernel function which makes use of the shape of the spectral response surrounding a spectral point. This is achieved by comparing the value of an spectral attribute in a spectrum and each of a set of neighbouring spectral attributes within a window around the spectral attribute. Weighting values can be applied to calculations when deriving the kernel function. The weighting values can assign different degrees of importance to different regions of the spectrum. The method can be used to: classify unknown spectra; predict the concentration of an analyte within a mixture; database searching for the closest match using a kernel-derived distance metric; visualisation of high-dimensional spectral data in two or three dimensions.07-15-2010
20100198758DATA CLASSIFICATION METHOD FOR UNKNOWN CLASSES - A system and method for creating a CD Tree for data having unknown classes are provided. Such a method can include dividing training data into a plurality of subsets of node training data at a plurality of nodes arranged in a hierarchical arrangement, wherein the node training data has a range. Furthermore, dividing node training data at each node can include, ordering the node training data, generating a plurality of separation points and a plurality of pairs of bins from the node training data, wherein each pair of bins includes a first bin and a second bin with a separation point being located between the first bin and the second bin, and classifying the node training data into either the first bin or the second bin for each of the separation points, wherein the classifying is based on a data classifier. Validation data can be utilized to calculate the bin accuracy between the node training data bin pairs and the validation data bin pairs for each separation point, and the separation point having a high bin accuracy can be selected as the node separation point.08-05-2010
20120197826INFORMATION MATCHING APPARATUS, METHOD OF MATCHING INFORMATION, AND COMPUTER READABLE STORAGE MEDIUM HAVING STORED INFORMATION MATCHING PROGRAM - The information matching apparatus includes: a training data rule setting unit that sets rules defining conditions for a training data of a positive example that is a pair of the records to be judged to be identical and a training data of a negative example that is a pair of the records to be judged to be non-identical; and a training data generating unit that, for the record of a matching source, generates a training data of the positive example by searching for the records of a matching target by using a positive example rule that is a rule defining conditions for the training data of the positive example, and generates a training data of the negative example by searching for the records of the matching target by using a negative example rule that is a rule defining conditions for the training data of the negative example.08-02-2012
20080208777METHODS AND APPARATUS FOR PREDICTIVE ANALYSIS - Methods and apparatus for predictive analytics generally comprise one or more artificial agents and an agent factory. An artificial agent may be responsive to at least one of an internal data set and an external data set. Further, an artificial agent may produce a correlation data set relating an outcome data set and at least one of the internal data set and the external data set. In addition, an artificial agent may produce a predictability value corresponding to the correlation data set. The agent factory may be responsive to the outcome data set. Also, the agent factory may produce the artificial agent in response to the outcome data set.08-28-2008
20100217730TEMPORALLY-CONTROLLED ITEM RECOMMENDATION METHOD AND SYSTEM BASED ON RATING PREDICTION - The present invention proposes a temporally-controlled item recommendation method and system based on rating prediction. According to this invention, the item recommendation method comprises inputting an item to be recommended; determining a temporal rating model related to the item, the temporal rating model being used to predict variation of the rating of the item with time; applying one or more recommendation strategies to the determined temporal rating model to determine optimal recommendation times of the item; and recommending the item to a user at the determined optimal recommendation times. In different embodiments, the temporal rating model of the item can be selected from a set of pre-stored temporal rating models or automatically generated according to history data in the system. In addition, the selected temporal rating model can be adjusted in accordance with user preference information or user feedback information. The item recommendation system of this invention is able to consider the change of a user's interest in a given item with time so as to increase the effectiveness of recommendations and improve user experience.08-26-2010
20100223216ARTIFICIAL VISION SYSTEM AND METHOD FOR KNOWLEDGE-BASED SELECTIVE VISUAL ANALYSIS - Generally the background of the present invention is the field of artificial vision systems, i.e. systems having a visual sensing means (e.g. a video camera) and a following processing stage implemented using a computing unit. The processing stage outputs a representation of the visually analysed scene, which output can then be fed to control different actors, such as e.g. parts of a vehicle (automobile, plane, . . . ) or a robot, preferably an autonomous robot such as e.g. a humanoid robot.09-02-2010
20100179932ADAPTIVE DRIVE SUPPORTING APPARATUS AND METHOD - Provided are an adaptive drive supporting apparatus and method that provide a personalized telematics user interface capable of supporting safe driving and convenient use. The adaptive drive supporting apparatus includes: a statistics database unit which stores and manages information on an average degree of attention required when a driving operation, a state of a car, or an external environment changes, information on degrees of attention required for manipulations of interfaces of the car, and a similarity between the functions of the interfaces; a personal characteristic setting unit which sets an individual degree of attention for each driver based on the average degree of attention according to a change in at least one of the driving operation, the state of the car, and the external environment; and an interface providing unit which determines whether or not a sum of the individual degree of attention and the degree of attention required when each driver manipulates a requested interface is larger than a predetermined threshold degree of attention required for safe driving.07-15-2010
20100228691Media Tag Recommendation Technologies - Technologies for recommending relevant tags for the tagging of media based on one or more initial tags provided for the media and based on a large quantity of other tagged media. Sample media as candidates for recommendation are provided by a set of weak rankers based on corresponding relevance measures in semantic and visual domains. The various samples provided by the weak rankers are then ranked based on relative order to provide a list of recommended tags for the media. The weak rankers provide sample tags based on relevance measures including tag co-occurrence, tag content correlation, and image-conditioned tag correlation.09-09-2010
20100228692SYSTEM AND METHOD FOR MULTI-MODAL BIOMETRICS - A system and method relate to multi-modal biometrics. A single modality score is generated for each of a plurality of biometric modalities. A classifier is selected from a database of multi-modal classifiers, and a multi-modal fusion is applied to the single modality scores using the classifier. The single modality scores are then aggregated. A context dependent model is generated, and a measure of the context in which the biometric samples were obtained is applied to the aggregated single modality scores. It is then determined whether there is a match between two or more biometric samples.09-09-2010
20100228693METHOD AND SYSTEM FOR GENERATING A DOCUMENT REPRESENTATION - A method, system and computer program product for generating a document representation are disclosed. The system includes a server and a client computer, and the method involves: receiving into memory a resource containing at least one sentence of text; producing a tree comprising tree elements indicating parts-of-speech and grammatical relations between the tree elements; producing semantic structures each having three tree elements to represent a simple clause (subject-predicate-object); and storing a semantic network of semantic structures and connections therebetween. The semantic network may be created from a user provided root concept. Output representations include concept maps, facts listings, text summaries, tag clouds, indices; and an annotated text. The system interactively modifies semantic networks in response to user feedback, and produces personal semantic networks and document use histories.09-09-2010
20100223213SYSTEM AND METHOD FOR PARALLELIZATION OF MACHINE LEARNING COMPUTING CODE - Systems and methods for parallelization of machine learning computing code are described herein. In one aspect, embodiments of the present disclosure include a method of generating a plurality of instruction sets from machine learning computing code for parallel execution in a multi-processor environment, which may be implemented on a system, of, partitioning training data into two or more training data sets for performing machine learning, identifying a set of concurrently-executable tasks from the machine learning computing code, assigning the set of tasks to two or more of the computing elements in the multi-processor environment, and/or generating the plurality of instruction sets to be executed in the multi-processor environment to perform a set of processes represented by the machine learning computing code.09-02-2010
20100274744System And Computer-Implemented Method For Generating Temporal Footprints To Identify Tasks - A system and computer-implemented method for generating temporal footprints to identify tasks is provided. One or more events performed by a user during execution of a task is recorded. Patterns including sequences of two or more of the events are identified. Each pattern occurs at a plurality of occurrences. A determination of whether each pattern is significant is made. A temporal distance between the events in each pattern occurrence for each pattern is identified. A pattern value is determined for each pattern based on a number of occurrences and the associated temporal distance. The pattern value is applied to a significance level. At least one of the patterns is determined to be significant when the pattern value satisfies the significance level. A temporal footprint is generated for the executed task and includes the significant patterns.10-28-2010
20120036094LEARNING APPARATUS, IDENTIFYING APPARATUS AND METHOD THEREFOR - A learning apparatus acquires a plurality of training samples containing a plurality of attributes and known classes, gives the plurality of training samples to a route node of a decision tree to be learned as an identifier, generates a plurality of child nodes from a parent node of the decision tree, allocates the training samples whose attribute corresponding to a branch condition for classification is not a deficit values at the parent node of the decision tree out of the plurality of training samples, to any of the plurality of child nodes according to the branch condition, gives the training samples whose attribute is the deficit value, to any one of the plurality of child nodes, and executes the generation of the child nodes and the allocating of the training samples until a termination condition is satisfied.02-09-2012
20100145892SEARCH DEVICE AND ASSOCIATED METHODS - A search device and associated methods use music emotions to browse, organize and retrieve music collections. The search device comprises a processor and an interface. The processor uses machine learning techniques to determine music emotion according to music features and organizes music by emotions for browsing and retrieving music collections. The interface connects to the processor and allows a person to retrieve desired music from the processor. Methods associated with the search device comprise a processor initialization method, a method of loading new music into the search device and several methods of retrieving desired music from the search device.06-10-2010
20100241597DYNAMIC ESTIMATION OF THE POPULARITY OF WEB CONTENT - Techniques are presented for estimating the current popularity of web content. Click and view data for articles are used to estimate popularity of the articles by analyzing click-through rates. Click-though rates are estimated such that a current click-through rate reflects fluctuations in popularity of articles through time.09-23-2010
20100235305SYSTEM AND METHOD OF ON-DEMAND DOCUMENT PROCESSING - A document processing method includes receiving, at a server with a network interface, electronic documents from a user. The server includes a software application adapted to recognize a class of electronic documents to which the electronic documents belong. The method also includes processing the electronic documents received from the user to extract data therefrom based on a recognition that the electronic documents belong to the class of electronic documents. The extracted data corresponds to a service being provided to the user. The method also includes automatically mapping the extracted data from the processed electronic documents to a data repository on the server. The data repository is accessible by the user through the network interface. The method also includes electronically generating output data based on the mapped data from the data repository to the user. The output data corresponds to the service being provided to the user.09-16-2010
20100138367SYSTEM, METHOD, AND PROGRAM FOR GENERATING NON-DETERMINISTIC FINITE AUTOMATON NOT INCLUDING e-TRANSITION - An initial setting unit receives from an input device a syntax tree generated from a regular expression, and initializes an NFA and an NFA converting section that applies five conversion patterns to each node of the syntax tree to directly convert the node into an NFA not including ε-transition. When the conversion is finished, the NFA converting section outputs the NFA generated to an output device.06-03-2010
20100138369LEARNING APPARATUS, LEARNING METHOD, INFORMATION MODIFICATION APPARATUS, INFORMATION MODIFICATION METHOD, AND PROGRAM - A content modification unit modifies an input image in accordance with a user operation, and generates modification information necessary for outputting a resulting output image. A modification information recording unit accumulates a plurality of pieces of modification information corresponding to the number of times an operation is performed by a user. A learning unit uses the plurality of pieces of modification information accumulated in the modification information recording unit as student data, and performs learning using teacher data acquired by a teacher data acquisition unit to calculate a prediction coefficient representing the feature of the user operation, and stores the prediction coefficient in a user algorithm recording unit. The present invention can be applied to, for example, an image processing apparatus.06-03-2010
20100250474PREDICTIVE CODING OF DOCUMENTS IN AN ELECTRONIC DISCOVERY SYSTEM - Embodiments of the invention relate to systems, methods, and computer program products for improved electronic discovery. More specifically, embodiments relate to computer program products for predictive and automated coding of identical or highly similar documents for the purpose of limiting the volume of documents requiring review and thereby increasing the overall efficiency of the document review process.09-30-2010
20100250473Active Learning Method for Multi-Class Classifiers - A method trains a multi-class classifier by iteratively performing the following steps until a termination condition is reached. The probabilities of class membership for unlabeled data obtained from an active pool of unlabeled data are estimated. A difference between a largest probability and a second largest probability is determined. The unlabeled data with the lowest difference is selected, labeled and then added to a training data set for training the classifier.09-30-2010
20110119208METHOD AND SYSTEM FOR DEVELOPING A CLASSIFICATION TOOL - An exemplary embodiment of the present invention provides a computer implemented method of developing a classifier. The method includes receiving input for a case, the case comprising a plurality of instances and an example, the example comprising a plurality of data fields each corresponding to one of the plurality of instances, wherein the input indicates which, if any, of the instances includes a data field belonging to a target class. The method also includes training the classifier based, at least in part, on the input from the trainer.05-19-2011
20100211534Efficient computation of ontology affinity matrices - In one embodiment, generating an ontology includes accessing an inverted index comprising a plurality of inverted index lists. An inverted index list may correspond to a term of a language. Each inverted index list may comprise a term identifier of the term and one or more document identifiers indicating one or more documents of a document set in which the term appears. The embodiment also includes generating a term identifier index according to the inverted index. The term identifier index comprises a plurality of sections and each section corresponds to a document. Each section may comprise one or more term identifiers of one or more terms that appear in the document.08-19-2010
20100211533EXTRACTING STRUCTURED DATA FROM WEB FORUMS - The web forum data extraction technique is designed for the structured data extraction of data on web forums using both page-level information and site-level knowledge. To do this, the technique finds the kinds of page objects a forum site has, which object a page belongs to, and how different page objects are connected with each other. This information can be obtained by re-constructing the sitemap of the target forum which is based on a Data Object Model of the target forum. The web forum data extraction technique collects three kinds of evidence for data extraction: 1) inner-page features which cover both semantic and layout information on an individual page; 2) inter-vertex features which describe linkage-related observations; and 3) inner-vertex features which characterize interrelationships among pages in one vertex. The technique employs Markov Logic Networks to combine the types of evidence statistically for inference and thereby can extract the desired structures.08-19-2010
20110022551METHODS AND SYSTEMS FOR GENERATING SOFTWARE QUALITY INDEX - Methods, systems and computer program code (software) products for generating a software quality index descriptive of quality of a given body of software code include identifying, by analysis of the body of software code, fault-prone files in the body of software code; constructing and training, by analysis of the body of software code, a model derived from analysis of the body of software code; and generating, based on the model, an index score representative of the quality of the body of software code.01-27-2011
20130218817ACTIVE ACQUISITION OF PRIVILEGED INFORMATION - A method for active learning using privileged information is disclosed. A processing device receives a set of labeled examples and a set of unlabeled examples. For each unlabeled example in the set of unlabeled examples, the processing device determines whether to query at least one of an oracle to obtain a label for the unlabeled example or a teacher to obtain privileged information about the unlabeled example. The processing device outputs a decision rule based on minimizing a number of queries to the oracle for a label and the teacher for privileged information. Minimizing the number of queries to the teacher and the oracle is based on a cost of querying the teacher or the oracle.08-22-2013
20100063947System and Method for Dynamically Adaptable Learning Medical Diagnosis System - A system and method for determining a likelihood of a disease presence in a particular patient includes a patient history database containing records. Each record includes a plurality of data fields related to a particular patient. An analyzing network is provided having access to the patient history database and having features based on the plurality of data fields included in the records to analyze the plurality of data fields and determine a likelihood of disease presence based on the plurality of features. A learning network is provided that has access to the analyzing network to review the likelihood of disease presence determined by the analyzing network and the plurality of data fields included in the records and automatically identify, evaluate, and add new features to the analyzing network that improve determinations of a likelihood of the disease.03-11-2010
20120143793FEATURE SPECIFICATION VIA SEMANTIC QUERIES - Technology is described that includes a method of feature specification via semantic queries. The method can include the operation of obtaining a data set having an identifier for each data row and a plurality of data features for each data row. A semantic query can be received that can be applied to the dataset that is usable by a machine learning tool. A entity feature map can be supplied that has entities and associated features for use by the machine learning tool. Further, a query structure can be analyzed using the entity feature map to identify input from the dataset for the machine learning tool.06-07-2012
20120143788TOXIN DETECTION SYSTEM AND METHOD - A system and method of generating a generic binary classifier for the presence of one or more toxins in water is provided. Features are extracted from a plurality of normalized a priori data sets that include one or more control data sets that are representative of an electric cell-substrate impedance sensor (ECIS) response to water with no toxins therein, and a plurality of treatment data sets that are representative of an ECIS response to water with a toxin therein. A plurality of classifier algorithms are trained using the extracted features, and a plurality of classification models are generated from each of the trained classifier algorithms. Each of the classification models is evaluated and, based on the evaluation of each classification model, a subset thereof is selected. The selected subset of the classification models is supplied as the generic binary classifier.06-07-2012
20120143798Electronic Communications Triage - Triaging electronic communications in a computing system environment can mitigate issues related to large volumes of incoming electronic communications. This can include an analysis of user-specific electronic communication data and associated behaviors to predict which communications a user is likely to deem important or unimportant. Client-side application features are exposed based on the evaluation of communication importance to enable the user to process arbitrarily large volumes of incoming communications.06-07-2012
20120143801INFORMATION CLASSIFICATION DEVICE, INFORMATION CLASSIFICATION METHOD, AND COMPUTER READABLE RECORDING MEDIUM - An information classification device (06-07-2012
20120143800Method and System for Knowledge Pattern Search and Analysis for Selecting Microorganisms Based on Desired Metabolic Property or Biological Behavior - Methods and systems for knowledge pattern search and analysis for selecting microorganisms based on desired metabolic properties or biological behaviors are disclosed in various embodiments of the invention. In one embodiment of the invention, a computer-implemented method for selecting a purpose-specific microorganism first compiles microorganisms' profiles by linking each microorganism's methanogenic, hydrogenic, electrogenic, another metabolic property, and/or another biological behavior to genetic and chemical fingerprints of metabolic and energy-generating biological pathways. Then, based on the compiled profiles of the microorganisms, the computer-implemented method groups the microorganisms into pathway characteristics using machine-learning and pattern recognition performed on a computer system, and subsequently generates a prediction called “discovered characteristics” for a desired metabolic property or a desired biological behavior of at least one microorganism. Furthermore, a profile match score may be calculated to indicate usefulness of one or more microorganisms for renewable energy generation from biological waste materials or wastewater.06-07-2012
20090259604METHODS, COMPUTER DEVICES, AND COMPUTER PROGRAM PRODUCTS FOR REGRESSION FROM INTERVAL TARGET VALUES - Methods, computing devices, and computer program products for regression from interval target values are provided. Training data having an interval output are read. An initial model is estimated. Representative values for the interval output are assigned using the initial model. A regression model is estimated using the representative values for the interval output. A determination is made whether the regression model converges. The step of assigning representative values for the interval output is iterated and the step of estimating the regression model using the representative values for the interval output iterated, in response to the regression model not converging. In response to the regression model converging, the regression model is output.10-15-2009
20100070439LEARNING SYSTEM AND LEARNING METHOD - A learning system according to the present invention includes an event list database for storing a plurality of event lists, each of the event lists being a set including a series of state-action pairs which reaches a state-action pair immediately before earning a reward, an event list managing section for classifying state-action pairs into the plurality of event lists for storing, and a learning control section for updating expectation of reward of a state-action pair which is an element of each of the event lists.03-18-2010
20110112994MUSICAL PIECE RECOMMENDATION SYSTEM, MUSICAL PIECE RECOMMENDATION METHOD, AND MUSICAL PIECE RECOMMENDATION COMPUTER PROGRAM - A musical piece recommendation system is provided that allows instantaneous registration of a new user and a new musical piece without retraining in a basic training section. A first incremental training section 05-12-2011
20100198761SYSTEMS, METHODS AND CIRCUITS FOR LEARNING OF RELATION-BASED NETWORKS - Circuits, devices and methods for processing learning networks are implemented using a variety of methods and devices. One example involves a circuit-implemented method to identify a relationship of objects in a set of objects. Local scores are generated for the object and possible parents. The local scores indicate relationship strength between object and parent. The results are stored in a memory. A state-machine circuit is used to perform sampling and searching of the parent sets for each data node. The local scores are used to encode orderings of the parent. An algorithm is executed that uses the encoded possible orderings and a random variable to generate and score a current order and a proposed order of the possible parent sets. The proposed orders are accepted or rejected based on probability rules applied to the scores for the current and proposed orders. Structures are sampled to assess a Bayesian-based relationship.08-05-2010
20100191682Learning Apparatus, Learning Method, Information Processing Apparatus, Data Selection Method, Data Accumulation Method, Data Conversion Method and Program - There is provided a learning apparatus including: a first data acquisition unit which acquires first user preference data belonging to a first data space; a second data acquisition unit which acquires second user preference data of a user in common with the first user preference data, the second user preference data belonging to a second data space which is different from the first data space; a compression unit which generates first compressed user preference data having less data item number from the first user preference data by utilizing a first set of parameters; and a learning unit which learns a second set of parameters utilized for generating second compressed user preference data having the same data item number as that of the first compressed user preference data from the second user preference data so that difference between the first compressed user preference data and the second compressed user preference data is to be small across a plurality of users.07-29-2010
20100191680METHOD AND APPARATUS FOR PREDICTING PREFERENCE RATING FOR CONTENT, AND METHOD AND APPARATUS FOR SELECTING SAMPLE CONTENT - Provided are a method and an apparatus for predicting a preference rating for content, and a method and an apparatus for selecting sample content in order to predict a preference rating for the content. In the method of predicting a preference rating for the content, a list of users having similar preferences to a target user is extracted from content usage information collected with respect to the same content, and the target user's preference rating for the content is predicted by applying preference rating information of the users with similar preferences to a machine learning algorithm.07-29-2010
20090281971SYSTEM AND METHOD FOR CLASSIFYING DATA STREAMS WITH VERY LARGE CARDINALITY - Systems and methods for object classification are provided. An object is identified along with the attributes that describe that object. These attributes are grouped into attribute patterns. Classes to be used in the classification are also identified. For each identified class a sketch table containing a plurality of parallel hash tables is created and trained using known objects with known classifications. For the object to be classified, each attribute pattern is processed using the all of the hash functions for each sketch table. This results in a plurality of values under each sketch table for a single attribute pattern. The lowest value is selected for each sketch table. The distribution of values across all sketch tables is evaluated for each attribute pattern. This produces a discriminatory power for each attribute pattern. Those attribute patterns having a discriminatory power above a given threshold are selected. The selected attribute patterns and associated sketch table values are added. The sketch table with the largest overall sum is identified, and the class associated with that sketch table is assigned to the object to which the attribute patterns belong.11-12-2009
20110112997INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing device includes a model learning unit that carries out learning for self-organization of internal states of a state transition prediction model which is a learning model having internal states, a transition model of the internal states, and an observation model where observed values are generated from the internal states, by using first time series data, wherein the model learning unit learns the observation model of the state transition prediction model after the learning using the first time series data, by fixing the transition model and using second time series data different from the first time series data, thereby obtaining the state transition prediction model having a first observation model where each sample value of the first time series data is observed and a second observation model where each sample value of the second time series data is observed.05-12-2011
20110112996Systems and methods for motion recognition using multiple sensing streams - Techniques for motion recognition using multiple data streams are disclosed. Multiple data streams from inertia sensors as well as non-inertial sensors are received to derive a motion recognition signal from motion recognizers. These motion recognizers are originally constructed from a training set of motion signals and may be updated with received multiple sensing signals. In one aspect, multiple data streams are converted to device-independent motion signals that are applied with the motion recognizers to provide a generalized motion recognition capability.05-12-2011
20090276382DETECTION OF UNKNOWN SCENARIOS - The present invention provides methods, systems and apparatus for detecting unknown scenarios in a data processing system. An example method includes the steps of: providing known scenario data describing one or more known scenarios in a database; generating element data depending on the known scenario data to form a set of elements, wherein each element is related to at least an actor and the behavior of the actor; computing subsets of elements by combining at least some of the elements of the set in dependence on their corresponding behavior; generating new scenario data related to new scenarios depending on the subsets of elements; and comparing the known scenario data with the new scenario data in order to identify the unknown scenarios.11-05-2009
20100191684Trainable hierarchical memory system and method - Memory networks and methods are provided. Machine intelligence is achieved by a plurality of linked processor units in which child modules receive input data. The input data are processed to identify patterns and/or sequences. Data regarding the observed patterns and/or sequences are passed to a parent module which may receive as inputs data from one or more child modules. the parent module examines its input data for patterns and/or sequences and then provides feedback to the child module or modules regarding the parent-level patterns that correlate with the child-level patterns. These systems and methods are extensible to large networks of interconnected processor modules.07-29-2010
20100191683CONDENSED SVM - The present invent ion provides a condensed SVM for high-speed learning using a large amount of training data. A first stage WS selector samples a plurality of training data from a training data DB, selects an optimal training vector x07-29-2010
20100191679METHOD AND APPARATUS FOR CONSTRUCTING A CANONICAL REPRESENTATION - Some embodiments provide systems and techniques to facilitate construction of a canonical representation (CR) which represents a logical combination of a set of logical functions. During operation, the system can receive a CR-size limit. Next, the system can construct a set of CRs based on the set of logical functions, wherein each CR in the set of CRs represents a logical function in the set of logical functions. The system can then combine a subset of the set of CRs to obtain a combined CR. Next, the system can identify a problematic CR which when combined with the combined CR causes the CR-size limit to be exceeded. The system can then report the problematic CR and/or a logical function associated with the problematic CR to a user, thereby helping the user to identify an error in the set of logical functions.07-29-2010
20090319452SYSTEM AND METHOD FOR AUTOMATICALLY LEARNING MAILBOX CONFIGURATION CONVENTIONS - A system and method automatically learns mailbox configuration conventions. The validator module determines a valid set of configuration parameters used for accessing an electronic mailbox of a user within a mail domain after receiving configuration information from the user that is limited in the configuration parameters required for accessing the electronic mailbox. A learner module accepts from the validator module a set of configuration parameters determined to be valid and generates configuration conventions for a mail domain. A database store is the generated configuration conventions. The validator and learner modules can be operative as part of a web server.12-24-2009
20080313109SYSTEM, METHOD AND COMPUTER PROGRAM PRODUCT FOR EVELUATING A STORAGE POLICY - A method for generating a storage policy, the method includes: receiving a storage system target function; and generating, by a machine learning entity, the storage policy in response to: (a) a set of file-related storage operation requests, (b) a state of the storage system before responding to the set of file-related storage operation requests, and (c) the storage system target function. A method for evaluating a storage policy, the method includes: simulating an application of the storage policy by the storage system during a first period, in response to a set of file-related storage operation requests that was provided to the storage system during the first period, to provide a simulation result; wherein the first period starts before the simulating.12-18-2008
20120143791METHOD AND APPARATUS FOR CAUSING AN APPLICATION RECOMMENDATION TO ISSUE - An apparatus may include a monitoring module configured to monitor user interactions by a user with applications. A contextual characteristics determiner may determine one or more contextual characteristics relating to the user interactions. Thereby, a data model builder may build a user behavior model for the user based at least in part on the user interactions and the contextual characteristics. The apparatus may provide for private storage of the user behavior module. A recommendation module may issue a recommendation, which may be mapped to one of the applications, based at least in part on the user behavior model. The recommendation may be issued in response to a query directed to a query module. The query may include current contextual characteristics of the user and/or the apparatus. The application recommendation may include one or more applications selected from one or more content providers, as controlled by a registrar module.06-07-2012
20090319451PATTERN CLASSIFICATION METHOD - For assigning a test pattern to a class chosen from a predefined set of classes, the class membership probability for the test pattern is calculated as well as the confidence interval for the class membership probability based upon a number of training patterns in a neighbourhood of the test pattern in the feature space. The number of training patterns in the neighbourhood of the test pattern is obtained from computing a convolution of a density function of the training patterns with a Gaussian smoothing function centred on the test pattern, where the density function of the training patterns is represented as a mixture of Gaussian functions. The convolution of the smoothing function and the mixture of Gaussian functions can be expressed analytically.12-24-2009
20090319448METHOD AND SYSTEM FOR POSITIONING - A positioning method includes: first, receiving wireless signals respectively at the positions of a number of training positions so as to extract a number of signal characteristics; next, establishing a positioning database according to the relationship between the training positions and the corresponding positioning module; then, classifying the training positions and the corresponding signal characteristics into a plurality of clusters, wherein when conducting positioning on a positioning node, a characteristic matching is conducted to find out a major cluster most similar to the positioning node; after that, conducting the characteristic matching between the positioning node and the training positions in the major cluster to decide a most-likely position of the positioning node. In addition, the present invention also provides a positioning system using the above-mentioned method.12-24-2009
20090112778Method and Apparatus for Leveraging End User Terminals in Self-Learning Networks - The invention includes a method and apparatus for configuring a self-learning network using feedback information received from an end user terminal communicating via the self-learning network. A method includes receiving feedback information from the end user terminal, generating configuration information for at least one network element of the self-learning network using the received feedback information, and configuring the at least one network element using the generated configuration information. The at least one network element of the self-learning network is configured by executing commands on each of the at least one network element and/or by propagating configuration information to each of the at least one network element. The feedback information may include user and/or terminal feedback information. The configuration information may include any information adapted for use in configuring the at least one network element of the self-learning network (and may also include configuration information for the end user terminal).04-30-2009
20100223215SYSTEMS AND METHODS OF MAKING CONTENT-BASED DEMOGRAPHICS PREDICTIONS FOR WEBSITES - Systems and methods for making demographic predictions for websites and web-pages. Embodiments include a system and a method of making demographic predictions for websites. The system and method select one or more websites with known demographic attributes for use as training websites, obtain demographic attributes data of the training websites, determine first features of web-pages of the training websites and develop a prediction model using the determined first features and the obtained demographic attributes data. The prediction model predicts one or more values for a target demographic attribute. The system and method determine second features of web-pages of a target website and apply the prediction model to the determined second features of the target website to predict one or more values for the target demographic attribute of the target website.09-02-2010
20110238605Information processing apparatus, information processing method, and program - An information processing apparatus including: a label acquisition section that acquires a label assigned by a user to a content selected among plural contents; a user certainty factor setting section that sets a user certainty factor to the label assigned by the user; a label prediction learning section that performs label prediction learning; a label prediction section that predicts a label regarding a content to which the label is not assigned, and calculates a label certainty factor that refers to certainty of the predicted label; a user certainty factor prediction section that performs user certainty factor prediction learning, and predicts a user certainty factor of (regarding) the predicted label of (regarding) the content to which the label is not assigned; and a selection section that selects a content to be next assigned a label among contents to which labels are not assigned.09-29-2011
20100223214AUTOMATIC EXTRACTION USING MACHINE LEARNING BASED ROBUST STRUCTURAL EXTRACTORS - A method and apparatus for automatically extracting information from a large number of documents through applying machine learning techniques and exploiting structural similarities among documents. A machine learning model is trained to have at least 50% accuracy. The trained machine learning model is used to identify information attributes in a sample of pages from a cluster of structurally similar documents. A structure-specific model of the cluster is created by compiling a list of top-K locations for each attribute identified by the trained machine learning model in the sample. These top-K lists are used to extract information from the pages of the cluster from which the sample of pages was taken.09-02-2010
20110060704DEPENDENCY GRAPH IN DATA-DRIVEN MODEL - The inference of a dependency graph that represents a graph of solves that leads from input model parameter(s) to output model parameters using analytics. In one embodiment, the dependency graph is part of visually driven analytics in which the output model parameter(s) are used to formulate data-drive scenes. As the identity of the input and/or output model parameter(s) change, or as the analytics themselves change, the dependency graph may also change. This might trigger a resolve of the analytics. In one embodiment, the intermediate parameters involved in the dependency graph may be viewed and evaluated by the user.03-10-2011
20120143799Method for Selecting Features Used in Continuous-Valued Regression Analysis - A method selects features used in continuous-valued regression analysis. Training data input to the method includes features and corresponding target values, wherein the target values are continuous, and there is one target value for each feature. Each threshold value is thresholded and discretized with respect to a threshold value to produce a discretized target value. Then, categorical feature selection is applied to the features, using the discrete target values, to produces selected features. The selected values can be used in any regression analysis.06-07-2012
20120143792PAGE SELECTION FOR INDEXING - Some implementations provide techniques for selecting web pages for inclusion in an index. For example, some implementations apply regularization to select a subset of the crawled web pages for indexing based on link relationships between the crawled web pages, features extracted from the crawled web pages, and user behavior information determined for at least some of the crawled web pages. Further, in some implementations, the user behavior information may be used to sort a training set of crawled web pages into a plurality of labeled groups. The labeled groups may be represented in a directed graph that indicates relative priorities for being selected for indexing.06-07-2012
20110112995Systems and methods for organizing collective social intelligence information using an organic object data model - A method for capturing and organizing intelligence data using an organic data model includes: receiving one or more webpages containing social intelligence data; segmenting content of the one or more webpages containing social intelligence data; identifying named entities in the segmented content of the one or more webpages; identifying topics in the segmented content of the one or more webpages; identifying opinions in the segmented content of the one or more webpages; integrating the identified named entities, topics, and opinions to construct an organic object data model; and storing organic object data associated with the constructed organic object data model in an organic object database.05-12-2011
20100318482Kernels for Identifying Patterns in Datasets Containing Noise or Transformation Invariances - Learning machines, such as support vector machines, are used to analyze datasets to recognize patterns within the dataset using kernels that are selected according to the nature of the data to be analyzed. Where the datasets include an invariance transformation or noise, tangent vectors are defined to identify relationships between the invariance or noise and the training data points. A covariance matrix is formed using the tangent vectors, then used in generation of the kernel, which may be based on a kernel PCA map.12-16-2010
20100299287Monitoring time-varying network streams using state-space models - In one embodiment, a statistical model is generated based on observed data, the observed data being associated with a network device, online parameter fitting is performed on parameters of the statistical model, and for each newly observed data value, a forecast value is generated based on the statistical model, the forecast value being a prediction of a next observed data value, a forecasting error is generated based on the forecast value and the newly observed data value, and whether the data of the network stream is abnormal is determined based on a log likelihood ratio test of the forecasting errors and a threshold value.11-25-2010
20110125681FEATURE EXTRACTION METHOD, FEATURE EXTRACTION APPARATUS, AND FEATURE EXTRACTION PROGRAM - Provided are a feature extraction method of creating a feature vector for objectively evaluating the sequence of aptamer on the basis of the biological features and a feature extraction apparatus and a feature extraction program for performing the method. The feature extraction method according to the present invention includes a step of predicting a secondary structure of a base sequence applied and a step of creating a feature vector based on a predicted secondary structure of the sequence.05-26-2011
20100306139CJK NAME DETECTION - Aspects directed to name detection are provided. A method includes generating a raw name detection model using a collection of family names and an annotated corpus including a collection of n-grams, each n-gram having a corresponding probability of occurring. The method includes applying the raw name detection model to a collection of semi-structured data to form annotated semi?structured data identifying n-grams identifying names and n?grams not identifying names and applying the raw name detection model to a large unannotated corpus to form a large annotated corpus data identifying n-grams of the large unannotated corpus identifying names and n-grams not identifying names. The method includes generating a name detection model, including deriving a name model using the annotated semi-structured data identifying names and the large annotated corpus data identifying names, deriving a not-name model using the semi?structured data not identifying names, and deriving a language model using the large annotated corpus.12-02-2010
20100312725SYSTEM AND METHOD FOR ASSISTED DOCUMENT REVIEW - A system and method for reviewing documents are provided. A collection of documents is portioned into sets of documents for review by a plurality of reviewers. For each set, documents in the set are displayed on a display device for review by a reviewer and temporarily organized through grouping and sorting. The reviewer's labels for the displayed documents are received. Based on the reviewer's labels, a class from a plurality of classes is assigned to each of the reviewed documents. A classifier model stored in computer memory is progressively trained, based on features extracted from the reviewed documents in the set and their assigned classes. Prior to review of all documents in the set, a calculated subset of documents for which the classifier model assigns a class different from the one assigned based on the reviewer's label is returned for a second review by a reviewer. Models generated from one or more other document sets can be used to assess the review of a first of the sets.12-09-2010
20090070279ESTIMATING THE EFFICACY OF A MATHEMATICAL MODEL OF SYSTEM BEHAVIOUR - Estimating the overall efficacy of a mathematical model of system behaviour involves providing a template representing factors that affect the overall efficacy of the mathematical model. A Bayesian Belief Network (BBN) having nodes based on the factors of the template is created and the BBN is used to obtain an estimate of the overall efficacy of the mathematical model of system behaviour.03-12-2009
20100325071SYSTEM AND METHOD FOR EMPIRICAL ENSEMBLE-BASED VIRTUAL SENSING OF GAS EMISSION - An empirical ensemble based virtual sensor system (VS) for the estimation of an amount of a gas (G) resulting from a combustion process (CP) comprising two or more empirical models (NN12-23-2010
20100312726FEATURE VECTOR CLUSTERING - One goal of computer services (e.g., email, web pages, blogs, advertisements, etc.) is to provide a user with Kinds (digital representations of everyday things) that may be relevant and interesting to the user. Users and Kinds may be plotted within a multidimensional matrix as feature vectors based upon their respective characteristics. An unsupervised clustering technique may be executed upon the matrix to create a mathematical cluster of feature vectors having similar characteristics. For example, a clothing cluster may comprise a dress Kind, a shoe Kind, a wool Kind, a watch Kind, etc. because the unsupervised clustering technique may determine these Kinds are plotted within the matrix in such a way that they have similar characteristics relating to clothing. The unsupervised clustering technique may also be utilized in determining which Kinds may be relevant to a user given a particular context with which a user is engaged with a computer resource.12-09-2010
20130138585Methods, Systems, And Computer Program Products For Recommending Applications Based On User Interaction Patterns - A method for recommending an application includes obtaining an input model representing user interaction patterns during execution of a first application. The input model is compared to a reference model representing user interaction patterns during execution of a second application. A similarity is determined between the input model and the reference model. A recommendation of the second application is generated in response to the similarity.05-30-2013
20130138586SERVICE GOAL INTERPRETING APPARATUS AND METHOD FOR GOAL-DRIVEN SEMANTIC SERVICE DISCOVERY - A service goal interpreting apparatus for goal-driven semantic service discovery is provided. The service goal interpreting apparatus includes a goal interpretation unit that interprets a goal of at least one service or application provided on the Web, and a goal registration unit that registers the goal interpreted by the goal interpretation unit in a service registry.05-30-2013
20130138588Identifying and ranking networked biographies and referral paths corresponding to selected qualifications - The most common automated search methods produce less-than-ideal results when searching online resumes, profiles, and the like (“biographies”) for the identities of people with a searcher-selected qualification (“candidates”). Keywords, their proximities, and their repetitions are less informative in biographies than in other informational documents. Similarly, chains of social connection (“referral paths”) do not always reveal the likelihood or ease of a searcher's introduction to a candidate. In both cases, the display order of results may be unrelated to any estimate of merit. To answer the question “Whom do I need and how do I reach them?” a classifier system uses heuristics or algorithms adapted to match the reactions of human experts on the selected qualifications. Terms in biographies, regardless of structure, are standardized and disambiguated for accurate comparisons, meaningful context is preserved, and biographies and referral paths are scored based on expected usefulness to the searcher.05-30-2013
20100325074REMOTE MONITORING THRESHOLDS - Apparatus for generating a threshold value indicative of a status change, comprising a trend projection engine for processing a plurality of sensed values in an order of value size to generate a corresponding predicted value for each of the plurality of sensed values, by reference to a value sequence trend of the plurality of sensed values, a comparator for comparing one or more of the plurality of sensed values against their corresponding predicted values to identify an abnormal sensed value which differs from its corresponding predicted value by a pre-specified amount, and a threshold generator for using the abnormal sensed value to identify the threshold value.12-23-2010
20100325070Isolating Changes in Dynamic Systems - A software optimization system isolates an effect of a change in a control variable from effects of ongoing, unknown changes in other variables. The system discards effects due to noise so that effects of interest to a programmer are more easily visible. The software optimization system treats variations in one or more control variables and in the output of the system as signals. The system varies the control variable at a specific frequency unlikely to correlate with uncontrolled variations in external variables. The system uses digital signal processing (DSP) techniques to filter the output, isolating the frequency of the control variable variation. The system then compares the resulting filtered output to the input to determine the approximate effect of the variation in the control variable.12-23-2010
20100325072SYSTEM AND METHOD FOR SOLVING MULTIOBJECTIVE OPTIMIZATION PROBLEMS - A system and method for solving a set of optimization problems initializes a current region of optimal solutions for the set of optimization problems, performs a reduction phase, and provides the optimal solutions within the current region. The reduction phase creates a random sample of points within the current region and identifies a subregion of the current region that very likely does not contain any optimal solutions. The identified subregion is then removed from the current region. If the current region does not satisfies one or more convergence criteria, the process loops back to create another random sample of points and repeats the above-described steps. If, however, the current region does satisfy the convergence criteria, the optimal solutions within the current region are provided to the output device.12-23-2010
20100332425Method for Clustering Samples with Weakly Supervised Kernel Mean Shift Matrices - A method clusters samples using a mean shift procedure. A kernel matrix is determined from the samples in a first dimension. A constraint matrix and a scaling matrix are determined from a constraint set. The kernel matrix is projected to a feature space having a second dimension using the constraint matrix, wherein the second dimension is higher than the first dimension. Then, the samples are clustered according to the kernel matrix.12-30-2010
20100332428ELECTRONIC DOCUMENT CLASSIFICATION - An electronic document classification system disclosed herein classifies electronic documents. The classification of the documents may involve analyzing the document and the information attached to the document to generate a set of classification data and comparing the classification data with one or more classification rules to generate a set of classifying data. The system attaches the set of classifying data to the electronic document and displays the electronic document based on the set of classifying data. The classification data may also be used to prioritize the electronic documents and to assign a retention period to the electronic documents. The system is further adapted to receive user feedback regarding the classification of the electronic document and to update the classification rules.12-30-2010
20100332424DETECTING FACTUAL INCONSISTENCIES BETWEEN A DOCUMENT AND A FACT-BASE - Techniques for identifying one or more inconsistencies between an unstructured document and a back-end fact-base are provided. The techniques include automatically parsing a query document and comparing the document with a back-end fact-base comprising facts relevant to the document, identifying one or more inconsistencies between information mentioned in the document and the facts stored in the back-end fact-base, and providing a response to the query document, wherein the response additionally includes the one or more identified inconsistencies.12-30-2010
20100332422Policy Evolution With Machine Learning - A method for constructing a classifier which maps an input vector to one of a plurality of pre-defined classes, the method steps includes receiving a set of training examples as input, wherein each training example is an exemplary input vector belonging to one of the pre-defined classes, learning a plurality of functions, wherein each function maps the exemplary input vectors to a numerical value, and determining a class for the input vector by combining numerical outputs of the functions determined for the input vector.12-30-2010
20100332427Data processing apparatus generating motion of 3D model and method - Provided is a data processing apparatus that may include a storage unit, a first calculator, and a second calculator. The storage unit may store a plurality of training data obtained by motion sensing. The first calculator may calculate a first transformation matrix by performing a regression analysis for the plurality of training data. The second calculator may calculate first output data by applying the first transformation matrix to first input data.12-30-2010
20100332426METHOD OF IDENTIFYING LIKE-MINDED USERS ACCESSING THE INTERNET - A method of identifying like-minded users accessing the Internet, comprises presenting a multimedia item to a user on an Internet site visited by the user. The user is offered at least one mechanism to provide a response to the item. Responses to the item from a plurality of users are collected. Those people providing the same response to the item are identified as like-minded.12-30-2010
20110010322METHOD AND SYSTEM FOR TRANSITIONING FROM A CASE-BASED CLASSIFIER SYSTEM TO A RULE-BASED CLASSIFIER SYSTEM - A computer implemented method including determining whether a predetermined condition is satisfied with respect to a case-based dataset stored within memory accessible by at least one processor; generating, for each of a plurality of rule-based classifiers, rule-based classification data identifying class boundaries between the records upon determining that the predetermined condition has been satisfied; computing a structural risk of the rule-based classification data for each of the rule-based classifiers with respect to the records identified within the record data; selecting rule-based classifiers having generated rule-based classification data identifying class boundaries between the records with a structural risk below a predetermined threshold; identifying selected rule-based classifiers having rule-based classification data that is within a predetermined degree of similarity with the case-based classification data; and replacing the case-based dataset stored within the memory with rule-based classification data of at least one of the identified rule-based classifiers.01-13-2011
20110010320Method, System, And Computer Program Product For Delivering Smart Services - A method, system, and computer program product are described for delivering smart services. According to an exemplary embodiment, a method for delivering smart services includes receiving a request to determine an availability of a service subscriber for responding to an event associated with a service. The service is defined in terms of the event and a situation of the service subscriber. A current situation of the service subscriber is determined using subscriber context information based on private information of the subscriber. Attributes of the event and the current subscriber situation are used to provide to the service at least one of the subscriber context information and a probability related to an availability of the subscriber for responding to the event, allowing the service to generate a response to the event on behalf of the subscriber without the service having direct access to the private subscriber information.01-13-2011
20110010317INFORMATION PROCESSING APPARATUS ENABLING DISCRIMINATOR TO LEARN AND METHOD THEREOF - An information processing apparatus includes a preliminary learning unit configured to learn a preliminary discriminator for a respective one of a plurality of combinations of variations in variation categories in a discrimination target pattern, a branch structure determination unit configured to perform discrimination processing using the preliminary discriminator and to determine a branch structure of a main discriminator based on a result of the discrimination processing, and a main learning unit configured to learn the main discriminator based on the branch structure.01-13-2011
20110016067PROBABILISTIC DECISION MAKING SYSTEM AND METHODS OF USE - Embodiments of this invention comprise modeling a subject's state and the influence of training scenarios, or actions, on that state to create a training policy. Both state and effects of actions are modeled as probabilistic using Partially Observable Markov Decision Process (POMDP) techniques. The POMDP is well suited to decision-theoretic planning under uncertainty. Utilizing this model and the resulting training policy with real world subjects creates a surprisingly effective decision aid for instructors to improve learning relative to a traditional scenario selection strategy. POMDP provides a more valid representation of trainee state and training effects, thus it is capable of producing more valid recommendations concerning how to structure training to subjects.01-20-2011
20110035345AUTOMATIC CLASSIFICATION OF SEGMENTED PORTIONS OF WEB PAGES - Exemplary methods and apparatuses are provided which may be used for classifying and indexing segmented portions of web pages and providing related information for use in information extraction and/or information retrieval systems.02-10-2011
20110029464SUPPLEMENTING A TRAINED MODEL USING INCREMENTAL DATA IN MAKING ITEM RECOMMENDATIONS - Incremental training data is used to supplement a trained model to provide personalized recommendations for a user. The personalized recommendations can be made by taking into account the user's behavior, such as, without limitation, the user's short and long term web page interactions, to identify item recommendations. A trained model is generated from training data indicative of the web page interaction data collected from a plurality of users. Incremental training data indicative of other web page interaction data can be used to supplement the trained model, or in place of the trained model. Incremental training data can be indicative of user behavior collected more recently than the data used to train the model, for example.02-03-2011
20110029466SUPERVISED RANK AGGREGATION BASED ON RANKINGS - A method and system for rank aggregation of entities based on supervised learning is provided. A rank aggregation system provides an order-based aggregation of rankings of entities by learning weights within an optimization framework for combining the rankings of the entities using labeled training data and the ordering of the individual rankings. The rank aggregation system is provided with multiple rankings of entities. The rank aggregation system is also provided with training data that indicates the relative ranking of pairs of entities. The rank aggregation system then learns weights for each of the ranking sources by attempting to optimize the difference between the relative rankings of pairs of entities using the weights and the relative rankings of pairs of entities of the training data.02-03-2011
20110029465DATA PROCESSING APPARATUS, DATA PROCESSING METHOD, AND PROGRAM - A data processing apparatus includes an obtaining unit configured to obtain time-series data from a wearable sensor, an activity model learning unit configured to learn an activity model representing a user activity state as a stochastic state transition model from the obtained time-series data, a recognition unit configured to recognize a current user activity state by using the activity model of the user obtained by the activity model learning unit, and a prediction unit configured to predict a user activity state after a predetermined time elapses from a current time from the current user activity state recognized by the recognition unit.02-03-2011
20110029463APPLYING NON-LINEAR TRANSFORMATION OF FEATURE VALUES FOR TRAINING A CLASSIFIER - A collection of labeled training cases is received, where each of the labeled training cases has at least one original feature and a label with respect to at least one class. Non-linear transformation of values of the original feature in the training cases is applied to produce transformed feature values that are more linearly related to the class than the original feature values. The non-linear transformation is based on computing probabilities of the training cases that are positive with respect to the at least one class. The transformed feature values are used to train a classifier.02-03-2011
20110035346METHOD AND SYSTEM FOR DATA ANALYSIS AND SYNTHESIS02-10-2011
20110035344COMPUTING MIXED-INTEGER PROGRAM SOLUTIONS USING MULTIPLE STARTING VECTORS - An optimization engine includes a mixed-integer programming (MIP) solver that receives a programming model, an outcome objective, and a group of start vectors. Each of the MIP start vectors in the group specify one or more restrictions to apply to the programming model. The MIP solver uses the programming model to compute a potential solution from each of the MIP start vectors included in the group, which results in a group of potential solutions. Next, the MIP solver selects one of the potential solutions in the group as an optimal intra-group solution. The optimal intra-group solution is the potential solution in the group that best achieves the outcome objective. In turn, the optimal intra-group solution is used to complete the outcome objective.02-10-2011
20110040711TRAINING A CLASSIFIER BY DIMENSION-WISE EMBEDDING OF TRAINING DATA - A classifier training method and apparatus for training, a linear classifier trained by the method, and its use, are disclosed. In training the linear classifier, signatures for a set of training samples, such as images, in the form of multi-dimension vectors in a first multi-dimensional space, are converted to a second multi-dimension space, of the same or higher dimensionality than the first multi-dimension space, by applying a set of embedding functions, one for each dimension of the vector space. A linear classifier is trained in the second multi-dimension space. The linear classifier can approximate the accuracy of a non-linear classifier in the original space when predicting labels for new samples, but with lower computation cost in the learning phase.02-17-2011
20110040712Extensions to Semantic Net - A semantic network includes a number of nodes are interconnected to one another through links (e.g., in a subject/verb/target form) representing relationships between the nodes and one or more of the links have one or more variants representing qualifications of the relationships between the nodes. For each link having one or more variants, the variants may be ordered in configurations. Such ordering of the variants in the configurations may be self-described within the semantic network and may determine precedence of those links belonging to the variants. Some of the links of the network may be nodes of others of the links. The interconnection of at least some of the nodes may define a meta-meta model that defines terms in which particular meta models can be defined, each meta model comprising meta facts regarding the nodes of the semantic network.02-17-2011
20110040707INTELLIGENT MUSIC SELECTION IN VEHICLES - A method of intelligent music selection in a vehicle includes learning user preferences for music selection in the vehicle corresponding to a plurality of driving conditions of the vehicle. Input is received that is indicative of a current driving condition of the vehicle. And, music is selected and played based on the learned user preferences for music selection in the vehicle corresponding to the current driving condition.02-17-2011
20110040706SCALABLE TRAFFIC CLASSIFIER AND CLASSIFIER TRAINING SYSTEM - A traffic classifier has a plurality of binary classifiers, each associated with one of a plurality of calibrators. Each calibrator trained to translate an output score of the associated binary classifier into an estimated class probability value using a fitted logistic curve, each estimated class probability value indicating a probability that the packet flow on which the output score is based belongs to the traffic class associated with the binary classifier associated with the calibrator. The classifier training system configured to generate a training data based on network information gained using flow and packet sampling methods. In some embodiments, the classifier training system configured to generate reduced training data sets, one for each traffic class, reducing the training data related to traffic not associated with the traffic class.02-17-2011
20110112993Search methods and various applications - The present invention relates to a system and method for information process using artificially constructed apparatus. More specially, in one preferred embodiment of the present invention, documents can be processed so that the most relevant terms of the contents of the documents can be obtained, and searched. In another preferred embodiment of the present invention, the present invention provides a system and method that can search for information in a document structure and provide precise results by analyzing the inputs and search results using the executing system and the knowledge structure of the think system.05-12-2011
20110055127MODEL OPTIMIZATION SYSTEM USING VARIABLE SCORING - A model optimization system is configured to determine quality of variables for model generation. A data storage stores input variables, quality metrics for the input variables, and weights for the quality metrics. The quality metrics describe sufficiency of data for the input variables and the data is provided for a plurality of regions. A scoring module determines a score for each region based on the input variables and the weighted quality metrics. An optimizer determines whether at least one of the input variables for a region is to be modified based on the scores, and determines whether the total score for the region is operable to be improved using a modified input variable.03-03-2011
20110055122MONITORING WITH ADAPTIVE DYNAMIC CLASSIFICATION - In a monitoring method, a time sequence of information pertaining to a monitored device, network, or system is recorded, comprising observations of the monitored device, network, or system and known prior correct action recommendations for the monitored device, network, or system. A hidden Markov model (HMM) operating on the time sequence of information is maintained. The HMM comprises a hidden state of the monitored device, network, or system. A current state of the monitored device, network, or system is classified using a classification value comprising an emission of the HMM that depends on an estimate of the distribution of the hidden state and on a selected portion of the time sequence of information. An action recommendation is generated for the current state of the monitored device, network, or system based on the classification value.03-03-2011
20090099986LEARNING TRADEOFFS BETWEEN DISCRIMINATIVE POWER AND INVARIANCE OF CLASSIFIERS - Systems and methods are described for learning the discriminative power-invariance tradeoffs for classification of input data (“tradeoff learning system”). In various embodiments, the tradeoff learning system receives multiple classifiers (“base classifiers”) and employs a learning technique to produce a combined classifier. Each received base classifier achieves a different level of tradeoff. The learning technique then decreases a function of kernel weights associated with each of the received classifiers to produce the combined classifier. By decreasing the function of kernel weights, the tradeoff learning system computes a combined classifier that classifies input data more accurately than the received multiple classifiers.04-16-2009
20090099985METHOD AND APPARATUS FOR IMPROVED REWARD-BASED LEARNING USING ADAPTIVE DISTANCE METRICS - The present invention is a method and an apparatus for reward-based learning of policies for managing or controlling a system or plant. In one embodiment, a method for reward-based learning includes receiving a set of one or more exemplars, where at least two of the exemplars comprise a (state, action) pair for a system, and at least one of the exemplars includes an immediate reward responsive to a (state, action) pair. A distance metric and a distance-based function approximator estimating long-range expected value are then initialized, where the distance metric computes a distance between two (state, action) pairs, and the distance metric and function approximator are adjusted such that a Bellman error measure of the function approximator on the set of exemplars is minimized. A management policy is then derived based on the trained distance metric and function approximator.04-16-2009
20100138370METHOD AND APPARATUS FOR MACHINE-LEARNING BASED PROFILING - A method and system for profiling a user based upon a user's previous on-line actions is provided. The profile provides a characterization of the user's preferences based upon a received user event. The user event identifying event identification information and a user identifier. A look-up in a cached web map is performed to retrieve classification information associated with the event identification information. A user profile is retrieved or created for the user identifier. Profile update information is generated based upon the retrieved classification information for the user event, to identify how the user is to be updated based upon the retrieved classification information and defined profiling rules. The user profile is updated and stored for access by an external advertising server. The classification information provides a text-score record comprising a text string and a score defined in relation to a lexical ontology comprising a hierarchy of categories.06-03-2010
20110213740SYSTEM AND METHOD FOR RESOURCE ADAPTIVE CLASSIFICATION OF DATA STREAMS - A system and method for resource adaptive classification of data streams. Embodiments of systems and methods provide classifying data received in a computer, including discretizing the received data, constructing an intermediate data structure from said received data as training instances, performing subspace sampling on said received data as test instances and adaptively classifying said received data based on statistics of said subspace sampling.09-01-2011
20110071967Automatic Labeler Assignment - A method, including receiving multi-labeler data that includes data points labeled by a plurality of labelers; building a model from the multi-labeler data, wherein the model includes an input variable that corresponds to the data points, a label variable that corresponds to true labels for the data points, and variables for the labels given by the labelers; and executing the model, in response to receiving new data points, to determine a level of expertise of the labelers for the new data points.03-24-2011
20110071964BUILDING AND USING PREDICTIVE MODELS OF CURRENT AND FUTURE SURPRISES - Methods are described for identifying events that would be considered surprising by people and identifying how and when to transmit information to a user about situations that they would likely find surprising. Additionally, the methods of identifying surprising situations can be used to build a case library of surprising events, joined with a set of observations before the surprising events occurred. Statistical machine learning methods can be applied with data from the case library to build models that can predict when a user will likely be surprised at future times. One or more models of context-sensitive expectations of people, a view of the current world, and methods for recording streams or events before surprises occur, and for building predictive models from a case library of surprises and such historical observations can be employed. The models of current and future surprises can be coupled with display and alerting machinery.03-24-2011
20090099983SYSTEM AND METHOD FOR AUTHORING AND LEARNING - Provided is a training method comprising delivering a situation to a learner, the situation comprising an event; automatically inserting a still menu in order to simulate a pause in the delivering; while pausing, recording an action by the learner in response to the event; reviewing the action recorded by the learner; and presenting to the learner a preferred action by a master in response to the event. Also provided is an authoring method comprising recording a situation comprising an event; further recording a master preferred action in response to the event; creating a program comprising motion menus and still menus where the situation, event, and master preferred action are placed on motion menus and still menus; and arranging the program and recordings so that the program carries out the following: delivering the recorded situation to the learner; automatically inserting a still menu after completing the play of a motion menu in order to simulate a pause in the delivering; recording the learner's action in response to the event; reviewing the learner's action; and presenting the learner with the master preferred action. An authoring and learning system is also provided, the system comprising a master video recording tool; a program comprising at least two motion menus and at least two still menus; a learner video delivery tool; a delivery medium; a learner video recording tool; and a learner control. The system may further comprise a program wherein the program gives an author flexibility to set up videos on alternating motion menus, and still menus and integrated recording and reviewing functions among the alternating menus, and wherein the program gives a learner some interactive control over the alternating menus and recording and reviewing functions.04-16-2009
20110125680Effects of Risk Factors on User Health - Risk factor data can be processed by a risk factor coaching engine to determine health risk for a user. The risk factor coaching engine may be executed within a health coaching protocol to perform actions that provide a user with information, recommendations and alerts via other coaching engines, and appointments with health care professionals. The risk factor coaching engine may also predict attribute values for a user based on a time period and goals for user health data upon which the predicted attribute value is based.05-26-2011
20100235306ADAPTIVE TIMELOG SYSTEM - An adaptive time log system that includes computer based systems and methods for monitoring, recording categorizing and reporting user activity on a timeline basis is provided.09-16-2010
20110125678GENERATING AN ACTIVITY INFERENCE MODEL FROM CONTEXTUAL DATA - One embodiment provides a system for generating an inference model that determines an activity type for a user from contextual information. During operation, the system receives a set of contextual information associated with the user, wherein the contextual information includes at least a set of location coordinates. The system then determines an association between the contextual information and an activity type. Next, the system generates an activity inference model based in part on the association, wherein the activity inference model takes an instance of contextual information as an input parameter and outputs a corresponding activity type. The model's parameters are based at least on statistics associated with the user's contextual history but not based on the complete contents of the user's contextual history.05-26-2011
20090099984SYSTEMS AND METHODS FOR GENERATING PREDICTIVE MATRIX-VARIATE T MODELS - Systems and methods predict missing elements from a partially-observed matrix by receiving one or more user item ratings; generating a model parameterized by matrices U, S, V; and outputting the model.04-16-2009
20100023465ACTIVE LEARNING SYSTEM, METHOD AND PROGRAM - A processing unit (01-28-2010
20110087625Systems and Methods for Automatic Creation of Agent-Based Systems - An agent-based system may be automatically generated from a specification provided by a user or third-party process. An agent generator may map the specification to a canonical model identifying one or more tasks to be performed by the agent-based system as ontological concepts. The agent generator may generate one or more candidate agents using the canonical model. The candidate agents may comprise one or more interconnected data transforms, which may comprise data access transforms, preprocessing transforms, machine learning transforms, and/or structural transforms. The agent generator iteratively modifies the agent-based system until a termination criteria is satisfied. The termination criteria may provide a selection mechanism whereby a performance of the plurality of candidate agents may be evaluated. An optimal agent may be selected using, inter alia, the performance of the agent-based system.04-14-2011
20100017349METHOD AND APPARATUS FOR DERIVING PROBABILISTIC MODELS FROM DETERMINISTIC ONES - A Dynamic Bayesian Network provides models that provides emulation of patient data.01-21-2010
20110093418AI Time Machine - A method for an AI time machine to accept sequential input tasks from at least one user, manage tasks, and execute tasks simultaneously or sequentially. Tasks specified by a user can be accomplished in the virtual world or in the real world and includes extracting digital data from electronic devices or manipulation of objects in the real world. The AI time machine's data structures, comprising: at least one dynamic robot to train the AI time machine; a main program with two modes: training mode and standard mode; external technologies, comprising: universal artificial intelligence programs, human level robots, psychic robots, super intelligent robots, the AI time machine, dynamic robots, a signalless technology, atom manipulators, ghost machines, a universal CPU, an autonomous prediction internet, and a 4-d computer; a videogame environment for virtual characters to do and store work; a prediction internet; a universal brain to store dynamic robot pathways or virtual character pathways, said universal brain, comprising: a real world brain, a virtual world brain, and a time machine world brain; a timeline of Earth that records predicted knowledge of Earth's past, current and future; a future United States government system; and a long-term memory. The present invention further serves as a universal AI to control at least one of the following: a machine, a hierarchical team of machines, a universal machine and a transforming machine.04-21-2011
20090048990Temporal Document Trainer and Method - An electronic document sorter is trained to classify documents based on their temporal qualities. The invention can be used in environments such as automated news aggregators, search engines and other electronic systems which compile information having temporal qualities.02-19-2009
20100042559METHOD AND APPARATUS FOR AUTOMATED IDENTIFICATION OF SIGNAL CHARACTERISTICS - A method of assessing a signal to identify particular signal characteristics comprises application of machine learning to multi-dimensional histograms derived from multi-tap sampling of the signal. The signal is sampled from at least two tap points to retrieve a sample set, and the at least two tap points are adapted to retrieve distinct samples from the signal, such as time spaced samples or spectrally distinct samples. Multiple sample sets are retrieved from the signal over time. The at least two dimensional histogram is built from the joint probability distribution of the plurality of sample sets. A machine learning algorithm then processes the multi-dimensional histogram, and is trained to predict a value of at least one characteristic of the signal.02-18-2010
20090313188Computationally Efficient Signal Classifier - Methods provided by this description may include receiving input signals for classification, and deriving specified signal parameters from the input signals. These methods may also compare the specified signal parameter to signal parameters derived from training signals, with the training signals being associated with predefined signal classes. These methods may also classify the input signals based on this comparison of the signal parameters, as derived respectively from the input signals in the training signals.12-17-2009
20090313190COGNITIVE OPERATING SYSTEM - An operating system configured to support cognitive capable environments is described. The system comprises a memory structure and an I/O process configured to update inputs and outputs. The operating system further includes a process to determine changes to symbols within the symbol space due to the stimulus and create STM images of the inputs. Additional processes are configured to pass stimulations between connecting symbols, measure temporal and spatial properties of images, filter the STM images based upon the properties, and propagate stimuli through hereditary structures. Further processes analyze propagated symbol groups of STM images for novel distinctions, create new symbols, connect novel stimuli together for symbol groupings, update existing symbols to include connections to the new symbols, and assign weights to the connections. Additional processes form stimulus-response pairs from received images, provide time-based erosion of connection weights of symbols and adapt learned responses to long term memories.12-17-2009
20100131438Medical Ontologies for Computer Assisted Clinical Decision Support - Medical ontology information is used for mining and/or probabilistic modeling. A domain knowledge base may be automatically or semi-automatically created by a processor from a medical ontology. The domain knowledge base, such as a list of disease associated terms, is used to mine for corresponding information from a medical record. The relationship of different terms with respect to a disease may be used to train a probabilistic model. Probabilities of a disease or chance of indicating the disease are determined based on the terms from a medical ontology. This probabilistic reasoning is learned with a machine from ontology information and a training data set.05-27-2010
20100131437Correlating data indicating subjective user states associated with multipleusers with data indicating objective occurrences - A computationally implemented method includes, but is not limited to acquiring subjective user state data including data indicating incidence of at least a first subjective user state associated with a first user and data indicating incidence of at least a second subjective user state associated with a second user; acquiring objective occurrence data including data indicating incidence of at least a first objective occurrence and data indicating incidence of at least a second objective occurrence; and correlating the subjective user state data with the objective occurrence data. In addition to the foregoing, other method aspects are described in the claims, drawings, and text forming a part of the present disclosure.05-27-2010
20100131436Soliciting data indicating at least one subjective user state in response to acquisition of data indicating at least one objective occurrence - A computationally implemented method includes, but is not limited to: acquiring objective occurrence data including data indicating occurrence of at least one objective occurrence; soliciting, in response to the acquisition of the objective occurrence data, subjective user state data including data indicating occurrence of at least one subjective user state associated with a user; acquiring the subjective user state data and correlating the subjective user state data with the objective occurrence data. In addition to the foregoing, other method aspects are described in the claims, drawings, and text forming a part of the present disclosure.05-27-2010
20120246100METHODS AND SYSTEMS FOR EXTRACTING KEYPHRASES FROM NATURAL TEXT FOR SEARCH ENGINE INDEXING - The present invention is a method and system for the extraction of keyphrases from natural text. For the purpose of this document, keyphrases are text segments that represent the main topic of a text. The method of the present invention may facilitate keyphrase extraction from any length of text. The text may be of several varieties, such as, for example a sentence, paragraph, document or collection of documents. Phrase separator methods may be applied to the text to extract phrases from the text. From these phrases the present invention may identify the one or more phrases that are integral to the meaning of the text and these may be identified as the keyphrases of the text. The text may be indexed using the keyphrases so that a search based upon any of the keyphrases will cause search engines and/or text retrieval means to retrieve the text.09-27-2012
20120246097Apparatus and Methods for Analyzing and Using Short Messages from Commercial Accounts - Disclosed are methods and apparatus for analyzing and using online short messages from promoting entity accounts (e.g., business or non-profit accounts). In one embodiment, a method of analyzing and using messages sent for a plurality of promoting entity accounts is disclosed. A plurality of models for classifying a plurality of messages based on a plurality of message features are obtained for each message. Each message is sent via a computer network between a selected one of the promoting entity accounts and one or more subscribing users that subscribe to receive messages from such selected promoting entity account, and each model is trained to identify whether a message belongs to a particular class based on a lexicon that was generated for such particular class and a training set of messages that belong to the particular class and message that do not belong to the particular class. A new message is classified based on the models and retaining classification information regarding the new message in a database that is accessible by a user so as to review the classification information on a computer display.09-27-2012
20090037351System and Method to Enable Training a Machine Learning Network in the Presence of Weak or Absent Training Exemplars - Described is a system and method for training a machine learning network. The method comprises initializing at least one of nodes in a machine learning network and connections between the nodes to a predetermined strength value, wherein the nodes represent factors determining an output of the network, providing a first set of questions to a plurality of users, the first set of questions relating to at least one of the factors, receiving at least one of choices and guesstimates from the users in response to the first set of questions and adjusting the predetermined strength value as a function of the choices/guesstimates. The real and simulated examples presented demonstrate that synthetic training sets derived from expert or non-expert human guesstimates can replace or augment training data sets comprised of actual training exemplars that are too limited in size, scope, or quality to otherwise generate accurate predictions.02-05-2009
20100036780MACHINE LEARNING - Computer implemented machine learning methods are described. A co-operative learning method involves a first rule based system and a second rule based system. A rule base is generated from input data and recursion data is used to recursively update the rule base as a result of newly received input data. Rule data defining at least one rule and associated data are sent to the second system which determines whether to update its rule base using the transmitted rule data, and if so the recursion data is used to recursively determine the updated rules for its rule base. A father machine learning method for a rule based system, involves receiving time series data, determining whether the data increases or decreases the spatial density for previously existing rules, and if so then creating a new cluster and associated rule, otherwise a new cluster is not created.02-11-2010
20100057647ACCOMMODATING LEARNED CLAUSES IN RECONFIGURABLE HARDWARE ACCELERATOR FOR BOOLEAN SATISFIABILITY SOLVER - A hardware accelerator is provided for Boolean constraint propagation (BCP) using field-programmable gate arrays (FPGAs) for use in solving the Boolean satisfiability problem (SAT). An inference engine may perform implications. Learned clauses may be generated during conflict analysis. Operations pertaining to learned clauses may include clause insertion and clause deletion (e.g., by invalidation) from a learned clause inference engine, and “garbage collection” in which unused or invalidated clauses may be removed from an inference engine.03-04-2010
20090216694MAXIMIZATION OF SUSTAINED THROUGHPUT OF DISTRIBUTED CONTINUOUS QUERIES - A system, method, and computer readable medium for optimizing throughput of a stream processing system are disclosed. The method comprises analyzing a set of input streams and creating, based on the analyzing, an input profile for at least one input stream in the set of input streams. The input profile comprises at least a set of processing requirements associated with the input stream. The method also comprises generating a search space, based on an initial configuration, comprising a plurality of configurations associated with the input stream. A configuration in the plurality of configurations is identified that increases throughput more than the other configurations in the plurality of configurations based on at least one of the input profile and system resources.08-27-2009
20090216692Information Processing Apparatus and Method, and Program - The invention relates to an information processing apparatus and method and a program which can provide an item suitable for a feeling (mood) of a user. A user data acquisition section 08-27-2009
20100070441Method, apparatus, and program for generating prediction model based on multiple regression analysis - An objective variable prediction model based on multiple regression analysis and having high prediction accuracy is generated by a computer. The method includes the steps of: a) constructing an initial sample set from samples whose measured value of an objective variable is known; b) obtaining a calculated value of the objective variable using multiple regression analysis; c) extracting samples whose difference between the measured and the calculated value is not larger than a first value, and calculating a determination coefficient by applying multiple regression analysis to the extracted samples; d) repeating the step c) by changing the first value until the determination coefficient exceeds a second value; and e) performing two-class classification to classify the sub-sample set obtained at the end of the step d) as a first sub-sample set and remaining samples as a second sub-sample set, and calculating a discriminant function.03-18-2010
20100063948MACHINE LEARNING METHODS AND SYSTEMS FOR IDENTIFYING PATTERNS IN DATA - Methods for training machines to categorize data, and/or recognize patterns in data, and machines and systems so trained. More specifically, variations of the invention relates to methods for training machines that include providing one or more training data samples encompassing one or more data classes, identifying patterns in the one or more training data samples, providing one or more data samples representing one or more unknown classes of data, identifying patterns in the one or more of the data samples of unknown class(es), and predicting one or more classes to which the data samples of unknown class(es) belong by comparing patterns identified in said one or more data samples of unknown class with patterns identified in said one or more training data samples. Also provided are tools, systems, and devices, such as support vector machines (SVMs) and other methods and features, software implementing the methods and features, and computers or other processing devices incorporating and/or running the software, where the methods and features, software, and processors utilize specialized methods to analyze data.03-11-2010
20100057651Knowledge-Based Interpretable Predictive Model for Survival Analysis - Knowledge-based interpretable predictive modeling is provided. Expert knowledge is used to seed training of a model by a machine. The expert knowledge may be incorporated as diagram information, which relates known causal relationships between predictive variables. A predictive model is trained. In one embodiment, the model operates even with a missing value for one or more variables by using the relationship between variables. For application, the model outputs a prediction, such as the likelihood of survival for two years of a lung cancer patient. A graphical representation of the model is also output. The graphical representation shows the variables and relationships between variables used to determine the prediction. The graphical representation is interpretable by a physician or other to assist in understanding.03-04-2010
20100057648CREATING FORMS WITH BUSINESS LOGIC - An eForm with integrated business logic is created from an existing eForm using a parser to parse a source eForm to extract attributes of items on the source eForm. An item recognition unit recognizes interactive items in the source eForm according to the attributes of the items extracted by the parser. A business logic recognition unit recognizes business logic integrated into the source eForm according to the attributes of the items. An object eForm generator generates an object eForm containing the recognized interactive items and business logic.03-04-2010
20100057649SYSTEM AND METHOD FOR FAULT PREDICTION IN HOME NETWORK - A system for fault prediction in a home network includes: a context generator for generating context information based on status data collected in real time about components of the home network; a specification interpreter for generating knowledge rules for fault detection by using specifications of the components of the home network; a context analyzer for analyzing if the context information meet the knowledge rules to classify the context information into normal situation contexts and abnormal situation contexts; a context pattern learner for generating new knowledge rules based on the abnormal situation contexts and fault rules corresponding to the abnormal situation contexts; a knowledge rule database for storing and managing the knowledge rules and the new knowledge rules; and a fault predictor for analyzing a correlation between the knowledge rules or the new knowledge rules and the generated context information, thereby predicting faults to be generated.03-04-2010
20110099133Systems and methods for capturing and managing collective social intelligence information - A method for capturing and managing training data collected online includes: receiving a first dataset from one or more online sources; sampling the first dataset and generating a second dataset, the second dataset including the data sampled from the first dataset; receiving an annotated second dataset with predefined labels; and dividing the annotated second dataset into a training dataset and a test dataset. The disclosed method further includes: configuring a machine learning based classifier based on the training dataset; predicting at least one data point based on the training dataset and calculating a confidence score; comparing the at least one predicted data point to the test dataset; sorting the at least one predicted data point based on its confidence score; and receiving corrected training data associated with the at least one predicted data point.04-28-2011
20110099135SYSTEM, METHOD AND COMPUTER PROGRAM PRODUCT FOR EVALUATING A STORAGE POLICY BASED ON SIMULATION - A computer implemented method for generating a storage policy for a storage system based on simulation results associated with a state of the storage system is provided. The method comprises receiving a target function applicable to a storage system, wherein the target function represents a measure of values associated with storage parameters related to productivity and loss tolerance of the storage system; wherein the simulation results for a state of the storage system are calculated based on a least one of (a) the storage system simulated response to a set of simulated file-related storage operation requests generated based on one or more simulation rules, (b) the state of the storage system before responding to the set of simulated file-related storage operation requests, (c) the storage system target function; and (d) rules for simulating file-related storage operation requests.04-28-2011
20100070438METHOD FOR PREDICTING INTERACTION BETWEEN PROTEIN AND CHEMICAL - The present invention has an object to provide a method for configuring a pattern recognizer using versatile, readily available data, comprehensive protein data, and comprehensive chemical data and an object to provide a method for predicting an unknown interaction of a pair by the pattern recognizer-configuring method. In particular, an interaction such as the coupling between a protein and a chemical is used as an index; at least one selected from four parameters that are the position of a peak in mass spectrum data obtained from each chemical, the set of the position and intensity of the peak, the distance between two peaks, and the set of the positions and intensities of the two peaks is vectorized for each of a first pair having a first interaction and a second pair having a second interaction; an amino acid sequence of each protein is vectorized; a vector containing elements of the vector derived from each protein and elements of the vector derived from each chemical paired with the protein is created; and a support vector machine (SVM) is applied to this vector and trained to learn them, whereby the pattern recognizer is configured so as to discriminate between a class to which the first pair belong and a class to which the second pair belong.03-18-2010
20100070437Information Management for Information Display Systems - A user-customizable information management solution, providing protection (e.g., privacy and/or security protection) for displayed information that may reduce or prevent exposure of sensitive and/or confidential information. In one aspect, a viewing aperture is controlled by the user and provides a view of a subset of the information displayed, where information not within the aperture is blocked or obscured to eliminate or reduce viewability. Optionally, simultaneous use of more than one viewing aperture may be supported. In another aspect, predefined information management instructions are used for determining how to protect a portion or portions of a document. The instructions may specify particular text and/or graphics categories defined by the user as being sensitive. Portions of the document that contain corresponding text and graphics are located, using a software-based search, and are blocked or obscured according to the predefined instructions. Dynamic tuning may be supported, whereby the user dynamically selects additional text/graphics for protecting.03-18-2010
20100070436METHOD AND APPARATUS FOR RECOMMENDING CONTENT ITEMS - A recommendation apparatus comprises a monitoring processor which monitors the presentation of content items. A sample processor determines preference data for different content items by performing the steps of determining a preference value for a content item presented by the presentation unit in response to a first duration for a first section of the content item being presented relative to a total duration of the content item, and if the first duration is less than the total duration, determining if a second section of the content item not being presented corresponds to at least one of an end section and a begin section of the content item; and if so determining a confidence value for the preference value in response to a second duration of the second section. The preference data is used as training data for determining a user preference model which is then used to generate recommendations.03-18-2010
20110106734SYSTEM AND APPARTUS FOR FAILURE PREDICTION AND FUSION IN CLASSIFICATION AND RECOGNITION - The present invention relates to pattern recognition and classification, more particularly, to a system and method for meta-recognition which can to predict success/failure for a variety of different recognition and classification applications. In the present invention, we define a new approach based on statistical extreme value theory and show its theoretical basis for predicting success/failure based on recognition or similarity scores. By fitting the tails of similarity or distance scores to an extreme value distribution, we are able to build a predictor that significantly outperforms random chance. The proposed system is effective for a variety of different recognition applications, including, but not limited to, face recognition, fingerprint recognition, object categorization and recognition, and content-based image retrieval system. One embodiment includes adapting machine learning approach to address meta-recognition based fusion at multiple levels, and provide an empirical justification for the advantages of these fusion element. This invention provides a new score normalization that is suitable for multi-algorithm fusion for recognition and classification enhancement.05-05-2011
20110099131PAIRWISE RANKING-BASED CLASSIFIER - The present invention provides methods and systems for binary classification of items. Methods and systems are provided for constructing a machine learning-based and pairwise ranking method-based classification model for binary classification of items as positive or negative with regard to a single class, based on training using a training set of examples including positive examples and unlabelled examples. The model includes only one hyperparameter and only one threshold parameter, which are selected to optimize the model with regard to constraining positive items to be classified as positive while minimizing a number of unlabelled items classified as positive.04-28-2011
20110099132PAY ZONE PREDICTION - Implementations of pay zone prediction are described. More particularly, apparatus and techniques described herein allow a user to predict pay zones in wells. By accurately predicting pay zones, the user can perforate an existing well at predefined well depths to access hydrocarbon bearing strata while avoiding other undesirable strata (such as water bearing strata). For example, in one possible implementation, well data and syntactic data from a first set of one or more existing wells can be used to create one or more syntactic models. These syntactic models can then be used with water cut and well data from the one or more existing wells to create a pay zone prediction model which can be used with wells outside of the first set of existing wells.04-28-2011
20110099130Integrated learning for interactive synthetic characters - A practical approach to real-time learning for synthetic characters grounded in the techniques of reinforcement learning and informed by insights from animal training. The approach simplifies the learning task for characters by (a) enabling them to take advantage of predictable regularities in their world, (b) allowing them to make maximal use of any supervisory signals, and (c) making them easy to train by humans. An autonomous animated dog is described that can be trained with a technique used to train real dogs called “clicker training.”04-28-2011
20100250472SYSTEM AND METHOD OF MACHINE-AIDED INFORMATION EXTRACTION RULE DEVELOPMENT - An automatic rule generation system generates rules for fact extraction. A rule generation module receives a sample and generates a rule from the sample. A rule relaxation module generates a relaxed rule from the rule. A rule testing module generates a reverse index from a corpus, applies the relaxed rule to the reverse index, and generates text segments. An information extraction module generates modified text segments from the relaxed rule and the text segments. A candidate suggestion module performs a candidate generation process: if the candidate generation process generates no candidates, the candidate suggestion module signals the rule relaxation module to generate a further relaxed rule to use as the relaxed rule. A user evaluates a candidate and provides the candidate as an additional sample for the automatic rule generation system to generate another rule to use as the rule. As a result of performing these actions iteratively, the rule is eventually generated and relaxed to result in an appropriate rule to use for fact extraction.09-30-2010
20110099134Method and System for Agent Based Summarization - A method and system for using a proxy agent based access to documents and the corresponding summaries and its subsequent usage is disclosed. The method and system provides for retrieving a document, generating or retrieving summary, generating statistical parameters to judge the summary quality, using text segmentation to judge the quality of the summary, getting user rating input and using it to train a classifier, using the classifier to predict the rating of a summary, displaying the summary along with its rating, and optionally overlaying the summary display with relevant advertising and thus prevent denial of information/information overload and stimulating accelerated learning.04-28-2011
20110153530Method and system for analyzing a legacy system based on trails through the legacy system - The present invention concerns a method for analyzing a legacy system (06-23-2011
20110153529METHOD AND APPARATUS TO EFFICIENTLY GENERATE A PROCESSOR ARCHITECTURE MODEL - A method and apparatus for efficiently generating a processor architecture model that accurately predicts performance of the processor for minimizing simulation time are described. In one embodiment, the method comprises: identifying a performance benchmark of a processor; sampling a portion of a design space for the identified performance benchmark; simulating the sampled portion of the design space to generate training data; generating a processor performance model from the training data by modifying the training data to predict an entire design space; and predicting performance of the processor for the entire design space by executing the processor performance model.06-23-2011
20110082819Systems and Methods for Decision Pattern Identification and Application - A system for decision pattern identification and application in a software engineering project includes a decision pattern miner configured to locate a plurality of decisions in a search space; a decision pattern generator configured to generate a decision pattern from the located decisions; a decision pattern repository configured to store the decision pattern; a decision pattern proposal maker configured to search the decision pattern repository for a decision pattern relevant to a decision space; and a decision pattern propagator configured to propagate the decision pattern relevant to the decision space in the decision space.04-07-2011
20110082820ASSESSING AN ENVIRONMENTAL FOOTPRINT OF AN OBJECT - In a method for assessing an environmental footprint of an object, economic data and environmental data of the object is aggregated and complexity of the object is determined. In addition, the economic data, the environmental data, and the complexity of the object are correlated and a model of an environmental footprint of the object is created based upon the correlation. Moreover, the environmental footprint of the object is assessed through application of the model.04-07-2011
20100125539Hybrid audio-visual categorization system and method - Meta-data (tags) for an audiovisual file can be generated by prompting a user to input certain tags (meta-data) descriptive of the audiovisual file, to serve as an initial estimate of the tags, and then revising the initial estimate (notably to expand it and/or render it more precise) based on the assumption that the relationships which hold between the different tags for a set of manually-tagged training examples will also hold for the tags of the input file now being tagged.05-20-2010
20130159220PREDICTION OF USER RESPONSE ACTIONS TO RECEIVED DATA - A system is provided for automatically predicting actions a user is likely to take in response to receiving data. The system may be configured to monitor and observe a user's interactions with incoming data and to identify patterns of actions the user may take in response to the incoming data. The system may enable a trainer component and a classifier component to determine the probability a user may take a particular action and to make predictions of likely user actions based on the observations of the user and the identified pattern of the user's actions. The system may also be configured to continuously observe the user's actions to fine-tune and adjust the identified patterns of user actions and to update the probabilities of likely user actions in order increase the accuracy of the predicted user action in response to incoming data.06-20-2013
20130159222INTERACTIVE INTERFACE FOR OBJECT SEARCH - Editorial curation of search results includes: receiving a search results page rendered in response to a search query; receiving user edits to the search results page, the user edits including changes to objects in the search results page; and applying the user edits to the search results page.06-20-2013
20130159224TECHNIQUES FOR REAL-TIME CUSTOMER PREFERENCE LEARNING - Techniques for real-time offer customer preference learning are presented. Local agents on communication channels are equipped with predefined rules that capture actions and behaviors of customers interacting with an enterprise. The metrics associated with these actions and behaviors are plugged into the rules and in some cases combined with known pre-existing preferences for the customers for purposes of evaluating the rules and creating newly learned preferences for the customers. The newly learned preferences are dynamically fed into offer evaluation processing to determine whether to make offers to the customers.06-20-2013
20110078097SHARED FACE TRAINING DATA - Face data sharing techniques are described. In an implementation, face data for a training image that includes a tag is discovered in memory on a computing system. The face data is for a training image that includes a tag associated with a face. The face data is replicated in a location in memory, on another computing system, so the face data is discoverable.03-31-2011
20110078098METHOD AND SYSTEM FOR EXTRACTION - A system and method for extracting information from at least one document in at least one set of documents, the method comprising: generating, using at least one ranking and/or matching processor, at least one ranked possible match list comprising at least one possible match for at least one target entry on the at least one document, the at least one ranked possible match list based on at least one attribute score and at least one localization score.03-31-2011
20110071966CONDITION MONITORING OF AN UNDERWATER FACILITY - A method for monitoring the condition of apparatus located at an underwater facility that includes sensing at least one parameter associated with the apparatus, providing a model of expected behaviour of said at least one parameter, comparing said sensed parameter with said model, and assessing the condition of the apparatus based upon said comparison.03-24-2011
20110071965SYSTEM AND METHOD FOR CROSS DOMAIN LEARNING FOR DATA AUGMENTATION - According to an example embodiment, a method comprises executing instructions by a special purpose computing apparatus to, for labeled source domain data having a plurality of original labels, generate a plurality of first predicted labels for the labeled source domain data using a target function, the target function determined by using a plurality of labels from labeled target domain data. The method further comprises executing instructions by the special purpose computing apparatus to apply a label relation function to the first predicted labels for the source domain data and the original labels for the source domain data to determine a plurality of weighting factors for the labeled source domain data. The method further comprises executing instructions by the special purpose computing apparatus to generate a new target function using the labeled target domain data, the labeled source domain data, and the weighting factors for the labeled source domain data, and evaluate a performance of the new target function to determine if there is a convergence.03-24-2011
20120303559CREATION, USE AND TRAINING OF COMPUTER-BASED DISCOVERY AVATARS - In embodiments of the present invention improved capabilities are described for developing, training, validating and deploying discovery avatars embodying mathematical models that may be used for document and data discovery and deployed within large data repositories.11-29-2012
20120303555Scalable Automatic Data Repair - A computer implemented method for generating a set of updates for a database comprising multiple records including erroneous, missing and inconsistent values, the method comprising using a set of partitioning functions for subdividing the records of the database into multiple subsets of records, allocating respective ones of the records to at least one subset according to a predetermined criteria for mapping records to subsets, applying multiple machine learning models to each of the subsets to determine respective candidate replacement values representing a tuple repair for a record including a probability of candidate and current values for the record, computing probabilities to select replacement values for the record from among the candidate replacement values which maximise the probability for values of the record for an updated database.11-29-2012
20120303557INTERACTIVE FRAMEWORK FOR NAME DISAMBIGUATION - A “Name Disambiguator” provides various techniques for implementing an interactive framework for resolving or disambiguating entity names (associated with objects such as publications) for entity searches where two or more same or similar names may refer to different entities. More specifically, the Name Disambiguator uses a combination of user input and automatic models to address the disambiguation problem. In various embodiments, the Name Disambiguator uses a two part process, including: 1) a global SVM trained from large sets of documents or objects in a simulated interactive mode, and 2) further personalization of local SVM models (associated with individual names or groups of names such as, for example, a group of coauthors) derived from the global SVM model. The result of this process is that large sets of documents or objects are rapidly and accurately condensed or clustered into ordered sets by that are organized by entity names.11-29-2012
20120303556COMPARISON OF MODELING AND INFERENCE METHODS AT MULTIPLE SPATIAL RESOLUTIONS - Embodiments provide a position service experimentation system to enable comparison of modeling and inference methods as well as characterization of input datasets for correspondence to output analytics. Crowd-sourced positioned observations are divided into a training dataset and a test dataset. A beacons model is generated based on the training dataset, while device position estimations are calculated for the test dataset based on the beacons model. The device position estimations are compared to the known position of the computing devices generating the positioned observations to produce accuracy values. The accuracy values are assigned to particular geographic areas based on the position of the observing computing device and aggregated to enable a systematic analysis of the accuracy values based on geographic area and/or positioned observations characteristics.11-29-2012
20120136815Display Device and Display Method - According to one embodiment, a display device includes an operation module, a display module, a recorder, a compiling module, and a display controller. The operation module receives an operation from a user. The display module displays content in accordance with the operation. The recorder measures the display time of the content that is being displayed and records the display time for each content. The compiling module compiles statistical information relating to a preference of the user based on the display time recorded for each content. The display controller displays, among contents that have been previously stored, content matching with the preference of the user indicated by the statistical information on the display module.05-31-2012
20120136817Data processing apparatus and method for motion synthesis - A data processing apparatus is used for motion synthesis. A preprocessing unit of the data processing apparatus calculates a mixture of factor analysis (MFA) parameter by applying an energy minimized optimization algorithm to motion capture data acquired in advance and stored in a motion database (DB). When a motion probability distribution model is generated as described above, a calculating unit of the data processing apparatus synthesizes a motion corresponding to input motion sensing data by applying the input motion sensing data to the motion probability distribution model.05-31-2012
20120136812METHOD AND SYSTEM FOR MACHINE-LEARNING BASED OPTIMIZATION AND CUSTOMIZATION OF DOCUMENT SIMILARITIES CALCULATION - One embodiment of the present invention provides a system for optimizing and customizing document-similarity calculation. During operation, the system presents a collection of similar documents to a user, collects feedback on the similarity of the documents from the user, generates generic rules for calculating document similarity, and filters documents with customized similarity calculation based on the feedback provided by the user.05-31-2012
20120136816NETWORK ANALYSIS SYSTEM - The present invention provides a method of operating a network comprising the steps of: analysing a first datastore comprising data representing historical network performance; creating or more indices within the first datastore; creating one or more probability networks in accordance with one or more of the created indices; determining from the one or more probability networks a conditional probability associated with an alarm event; and' if the conditional probability determined is less than a threshold value, disregarding the associated alarm event; or if the conditional probability determined is greater than a threshold value, using the associated alarm event in conjunction with other historical network data to predict future alarm events.05-31-2012
20120136814MUSIC RECOMMENDATION METHOD AND APPARATUS - A music recommendation method may include obtaining the music belongingness function of music, which is the set of granularity of music in different dimensions, wherein the dimension is the classification of music and the granularity is the classification of the dimension; obtaining the user belongingness function of a user, which is the set of granularity indicating likes of user in different dimensions; calculating a granularity correlation function by using the music belongingness function and the user belongingness function; calculating the value of the probability function indicating likes of user for music by using the granularity correlation function and a dimension weighting coefficient; and recommending the music to the user when the value of the probability function indicating likes of user for music is greater than a preset threshold. An apparatus applying to the method comprises corresponding modules.05-31-2012
20120136813METHOD OF PATTERN RECOGNITION IN A SIGNAL - The invention is directed to a method for pattern recognition in a signal corresponding for example to the steering angle of a vehicle for testing tires. The method comprises three major steps, namely step a) consisting in identifying phases in the signal by detecting phase changes; step b) consisting in classifying at least some of the identified phases based on their shapes and step c) consisting in detecting the presence of predetermined patterns in the signal where each predetermined pattern corresponds to a specific sequence of classes of phases. The phase changes are determined by extrema of the signal and its first derivative. The classification of the phase is made by means of parameters of the phases, namely the length dL, the amplitude dH, and a form factor S. The definition of the different classes is adjusted in a parameter space by means of manual recognition of maneuvers.05-31-2012
20110251980Interactive Optimization of the Behavior of a System - An interactive tool is described for modifying the behavior of a system, such as, but not limited to, the behavior of a classification system. The tool uses an interface mechanism to present a current global state of the system. The tool accepts one or more refinements to this global state, e.g., by accepting individual changes to parameter settings that are presented by the interface mechanism. Based on this input, the tool computes and displays the global implications of the updated parameter settings. The process of iterating over one or more cycles of user updates, followed by computation and display of the implications of the attempted refinements, has the effect of advancing the system towards a global state that exhibits desirable behavior.10-13-2011
20110060703METHOD AND SYSTEM FOR DETECTING CORRELATION IN DATA SETS - A method and system for detecting correlations in a data set is provided. The method includes determining one or more parameters associated with one or more data sets. The one or more parameters are determined at runtime for generating one or more test data sets from the one or more data sets. A test data of the one or more test data sets comprises one or more objects and one or more indices. The one or more objects are associated with the one or more indices. The method further includes computing one or more correlation coefficients associated with the one or more test data sets. The one or more correlation coefficients are computed for detecting correlation corresponding to the one or more test data sets.03-10-2011
20110060709DATA PROCESSING APPARATUS, DATA PROCESSING METHOD, AND PROGRAM - A data processing apparatus includes an action learning unit configured to train a user activity model representing activity states of a user in the form of a probabilistic state transition model using time-series location data items of the user, an action recognizing unit configured to recognize a current location of the user using the user activity model obtained through the action learning unit, an action estimating unit configured to estimate a possible route for the user from the current location recognized by the action recognizing unit and a selection probability of the route, and a travel time estimating unit configured to estimate an arrival probability of the user arriving at a destination and a travel time to the destination using the estimated route and the estimated selection probability.03-10-2011
20110060708INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing device includes: an object module determining unit for determining of a learning model having a time series pattern storage model for storing a time series pattern as a module which is the minimum component, a maximum likelihood module having the maximum likelihood, or a new module to be an object module that is a module having a model parameter of the storage model to be updated; and an updating unit for updating the model parameter of the object module using learned data to be used for learning that is the time series of an observed value; with the object module determining unit using the learned data to determine the maximum likelihood module or the new module to be the object module based on the posterior probability of the learning model in the case that learning of the maximum likelihood module or the new module has been performed.03-10-2011
20100325073NITROGEN OXIDE SENSITIVE FIELD EFFECT TRANSISTORS FOR EXPLOSIVE DETECTION COMPRISING FUNCTIONALIZED NON-OXIDIZED SILICON NANOWIRES - An apparatus for detecting volatile compounds derived from explosive materials with very high sensitivity. The apparatus is composed of field effect transistors of non-oxidized silicon nanowires modified with specific functional groups including, in particular, amine, imine and/or carboxyl moieties. Further a system is provided comprising the apparatus in conjunction with learning and pattern recognition algorithms and methods of use thereof for detecting and quantifying specific explosive compounds.12-23-2010
20110040710STRUCTURED DIFFERENTIAL LEARNING - A method and information processing system train a control system using structured differential learning. A set of features extracted from a set of input data is analyzed by a plurality of analyzing components. An output response is generated by each analyzing component in the plurality of analyzing components for each feature regarding whether the each feature has an acceptable value associated therewith relative to a value of the parameter associated with the each analyzing component. A confidence score is associated with the each output response. Each output response and each confidence score is combined into a single final output response and single final confidence score. An analyzing component is identified from the plurality of analyzing components that is a strongest candidate for generating an incorrect final output response based at least on the confidence score.02-17-2011
20110040709PATTERN BEHAVIOR SUPPORT IN A RULE ENGINE - Some embodiments of pattern behavior support in a rule engine have been presented. In one embodiment, a behavior builder registry is stored on a computer-readable storage device in a server. The behavior builder registry includes a set of behaviors supported by a rule engine and a set of builders associated with the behaviors. A compiler running on the server may compile a set of rules using the behavior builder registry to generate a network having a set of nodes. In response to a data object asserted propagating into a node of the network, the rule engine may first check one or more behaviors at the node before applying one or more regular constraints at the node on the data object asserted.02-17-2011
20110040708MULTIPLE ENTRY POINT NETWORK FOR STREAM SUPPORT IN A RULE ENGINE - Some embodiments of a multiple entry point network for stream support in an exemplary rule engine have been presented. In one embodiment, a stream of events is asserted into a working memory of a rule engine, which supports event processing. The rule engine, running on a server, processes the stream of events against a set of rules retrieved from a rule repository of the rule engine. To process the events, the rule engine may construct a network having multiple root nodes, each being an entry point into the network, through which the events may enter the network and propagate through the network.02-17-2011
20130166481DISCRIMINATIVE DECISION TREE FIELDS - A tractable model solves certain labeling problems by providing potential functions having arbitrary dependencies upon an observed dataset (e.g., image data). The model uses decision trees corresponding to various factors to map dataset content to a set of parameters used to define the potential functions in the model. Some factors define relationships among multiple variable nodes. When making label predictions on a new dataset, the leaf nodes of the decision tree determine the effective weightings for such potential functions. In this manner, decision trees define non-parametric dependencies and can represent rich, arbitrary functional relationships if sufficient training data is available. Decision trees training is scalable, both in the training set size and by parallelization. Maximum pseudolikelihood learning can provide for joint training of aspects of the model, including feature test selection and ordering, factor weights, and the scope of the interacting variable nodes used in the graph.06-27-2013
20110213738METHODS AND APPARATUS TO MODEL END-TO-END CLASS OF SERVICE POLICIES IN NETWORKS - Methods and apparatus to model end-to-end class of service policies in operational networks are disclosed. An example method to generate a class of service model is described, including electronically generating a ruleset based on the class of service configuration associated with a router, electronically generating a flat representation of the ruleset, electronically generating a class of service model by composing the flat representation into a composed ruleset, and storing the class of service model in a computer-readable memory.09-01-2011
20110213737TRAINING AND VERIFICATION USING A CORRELATED BOOSTED ENTITY MODEL - A system, method and program product training and verifying using an identity or entity model. A training system is disclosed that includes: a feature correlation system that groups features from an inputted feature data sample into subsets; a plurality of classifiers that determine if each feature classifies into an associated one of a plurality of feature models that make up the entity model; and a boosting system that boosts features from a subset for a next round of training if any of the features classify and at least one correlated feature from the subset does not classify. A verification system is disclosed that includes an identity model for the entity comprising a plurality of feature models, wherein each feature model is utilized to model a unique feature; a system for receiving a feature data sample and partitioning the feature data sample into a plurality of features; a system for determining if each of the plurality of features classifies into an associated feature model; and a voting system for analyzing a result of each attempted classification and determining an overall verification result.09-01-2011
20110213736METHOD AND ARRANGEMENT FOR AUTOMATIC CHARSET DETECTION - The invention relates, in an embodiment, to a method for handling a received document. The method includes receiving a plurality of text document samples. The method includes training, using a plurality of text document samples, to obtain a set of machine learning models. Training includes generating fundamental units from the plurality of text document samples for charsets of the plurality of text document samples. Training includes extracting a subset of said fundamental units as feature lists and converting the feature lists into a set of feature vectors. Training further includes generating the set of machine learning models from the set of feature vectors. The method includes applying the set of machine learning models against a set of target document feature vectors converted from the received document. The method includes decoding the received document to obtain decoded content of the received document based on at least the first encoding scheme.09-01-2011
20110258152CATEGORIZATION AUTOMATION - A method for categorization using multiple categories including obtaining multiple uniform resource locators (URLs) associated with the multiple categories, collecting multiple web pages identified by the multiple URLs, generating vocabulary terms based on the multiple web pages, generating an N-gram file including the multiple vocabulary terms, generating multiple classified URLs by labeling the plurality of URLs based on the multiple categories, generating multiple feature vectors by processing the classified URLs and the multiple web pages against the N-gram file, generating a categorization model by applying a machine learning algorithm to the multiple feature vectors, and loading a classifier with the categorization module and the N-gram file.10-20-2011
20130159219Predicting the Likelihood of Digital Communication Responses - Different advantageous embodiments provide for response prediction. A social element is received by a prediction mechanism. A feature set is generated for the social element. A prediction is generated using the feature set and a prediction model.06-20-2013
20120123978Learning Tags for Video Annotation Using Latent Subtags - A tag learning module trains video classifiers associated with a stored set of tags derived from textual metadata of a plurality of videos, the training based on features extracted from training videos. Each of the tag classifiers is comprised of a plurality of subtag classifiers relating to latent subtags within the tag. The latent subtags can be initialized by clustering cowatch information relating to the videos for a tag. After initialization to identify subtag groups, a subtag classifier can be trained on features extracted from each subtag group. Iterative training of the subtag classifiers can be accomplished by identifying the latent subtags of a training set using the subtag classifiers, then iteratively improving the subtag classifiers by training each subtag classifier with the videos designated as conforming closest to that subtag.05-17-2012
20120150771Knowledge Corroboration - Knowledge corroboration is described. In an embodiment many judges provide answers to many questions so that at least one answer is provided to each question and at least some of the questions have answers from more than one judge. In an example a probabilistic learning system takes features describing the judges or the questions or both and uses those features to learn an expertise of each judge. For example, the probabilistic learning system has a graphical assessment component which aggregates the answers in a manner which takes into account the learnt expertise in order to determine enhanced answers. In an example the enhanced answers are used for knowledge base clean-up or web-page classification and the learnt expertise is used to select judges for future questions. In an example the probabilistic learning system has a logical component that propagates answers according to logical relations between the questions.06-14-2012
20090012922METHOD AND APPARATUS FOR REWARD-BASED LEARNING OF IMPROVED SYSTEMS MANAGEMENT POLICIES - In one embodiment, the present invention is a method for reward-based learning of improved systems management policies. One embodiment of the inventive method involves supplying a first policy and a reward mechanism. The first policy maps states of at least one component of a data processing system to selected management actions, while the reward mechanism generates numerical measures of value responsive to particular actions (e.g., management actions) performed in particular states of the component(s). The first policy and the reward mechanism are applied to the component(s), and results achieved through this application (e.g., observations of corresponding states, actions and rewards) are processed in accordance with reward-based learning to derive a second policy having improved performance relative to the first policy in at least one state of the component(s).01-08-2009
20100280980System and Method for Resolving Gamma Ray Spectra - A system for identifying radionuclide emissions is described. The system includes at least one processor for processing output signals from a radionuclide detecting device, at least one training algorithm run by the at least one processor for analyzing data derived from at least one set of known sample data from the output signals, at least one classification algorithm derived from the training algorithm for classifying unknown sample data, wherein the at least one training algorithm analyzes the at least one sample data set to derive at least one rule used by said classification algorithm for identifying at least one radionuclide emission detected by the detecting device.11-04-2010
20100280979MACHINE LEARNING HYPERPARAMETER ESTIMATION - A method of determining hyperparameters (HP) of a classifier (11-04-2010
20100114804REPRESENTATIVE HUMAN MODEL GENERATION METHOD - A representative human model generation method is provided. The method includes i) setting a design target population, ii) setting a target accommodating percentage of the design target population, iii) converting anthropometric sizes of the design target population to normalized squared distances, iv) setting a boundary region for a target accommodation percentage of the design target population using normalized squared distances, and v) forming a minimum number of clusters satisfying the target accommodation percentage by performing cluster analysis for anthropometric cases contained in the boundary region among the design target population.05-06-2010
20110153528PROVIDING COMPARISON EXPERIENCES IN RESPONSE TO SEARCH QUERIES - Computer-readable media, computer systems, and computing devices facilitate providing a comparison experience to a user in response to a search query. Upon receiving a search query from the user, entities are extracted from the query. The entities are associated with entity classes. The entities, entity classes, previous user behavior, and other information are used to infer whether the user likely is engaging in a comparison task. If the inference indicates that the user likely is engaging in a comparison task, a comparison experience is generated and access to the comparison experience is provided to the user.06-23-2011
20110161258Method for Converting Dynamical Systems with Continuous States into Markov Decision Processes with Discrete States - A continuous dynamical system is converted to a Markov decision process (MDP) with discrete states. A predetermined number of continuous states of the continuous system is selected, wherein each continuous state corresponds to one discrete state of the MDP. Delaunay triangulation is applied to the continuous states to produce a set of triangles, wherein vertices of each triangle represent the continuous states. For each discrete state, a next discrete state y=f(x, a) is determined, wherein x represents the continuous state corresponding to the discrete state, a is a control action, and f is a non-linear transition function for the continuous. A particular triangle containing the next discrete state y is identified, and the next discrete state y is expressed as probabilities of transitioning to the discrete states corresponding to the continuous states x represented by the vertices of the particular triangle.06-30-2011
20100100513METHOD AND APPARATUS FOR PROVIDING FAST KERNEL LEARNING ON SPARSE DATA - A method and apparatus based on transposition to speed up learning computations on sparse data are disclosed. For example, the method receives an support vector comprising at least one feature represented by one non-zero entry. The method then identifies at least one column within a matrix with non-zero entries, wherein the at least one column is identified in accordance with the at least one feature of the support vector. The method then performs kernel computations using successive list merging on the at least one identified column of the matrix and the support vector to derive a result vector, wherein the result vector is used in a data learning function.04-22-2010
20120203720ROBUST PATTERN RECOGNITION SYSTEM AND METHOD USING SOCRATIC AGENTS - A computer-implemented pattern recognition method, system and program product, the method comprising in one embodiment: creating electronically a linkage between a plurality of models within a classifier module within a pattern recognition system such that any one of said plurality of models may be selected as an active model in a recognition process; creating electronically a null hypothesis between at least one model of said plurality of linked models and at least a second model among said plurality of linked models; accumulating electronically evidence to accept or reject said null hypothesis until sufficient evidence is accumulated to reject said null hypothesis in favor of one of said plurality of linked models or until a stopping criterion is met; and transmitting at least a portion of the electronically accumulated evidence or a summary thereof to accept or reject said null hypothesis to a pattern classifier module.08-09-2012
20120203719AUDIO SIGNAL PROCESSING DEVICE, AUDIO SIGNAL PROCESSING METHOD, AND PROGRAM - An audio signal processing device includes: a time-frequency analysis unit performing a time-frequency analysis of an input audio signal; a base factorization unit inputting learning data that is generated in advance based on an audio signal for learning including a sound from a plurality of sound sources and is made with base frequencies corresponding to the respective sound sources and carrying out base factorization of a time-frequency analysis result to the input audio signal inputted from the time-frequency analysis unit by applying a total base frequency that has the base frequencies corresponding to the respective sound sources combined therein to generate a base activity to the input audio signal; and a command identification unit inputting the base activity from the base factorization unit to carry out command identification by performing an identification process of the inputted base activity.08-09-2012
20120203718ALGORITHM ENGINE FOR USE IN A PATTERN MATCHING ACCELERATOR - A pattern matching accelerator (PMA) for assisting software threads to find the presence and location of strings in an input data stream that match a given pattern. The patterns are defined using regular expressions that are compiled into a data structure comprised of rules subsequently processed by the PMA. The patterns to be searched in the input stream are defined by the user as a set of regular expressions. The patterns to be searched are grouped in pattern context sets. The sets of regular expressions which define the pattern context sets are compiled to generate a rules structure used by the PMA hardware. The rules are compiled before search run time and stored in main memory, in rule cache memory within the PMA or a combination thereof. For each input character, the PMA executes the search and returns the search results.08-09-2012
20120203717Learning Similarity Function for Rare Queries - Techniques are described for determining queries that are similar to rare queries. An n-gram space is defined to represent queries and a similarity function is defined to measure the similarities between queries. The similarity function is learned by leveraging training data derived from user behavior data and formalized as an optimization problem using a metric learning approach. Furthermore, the similarity function can be defined in the n-gram space, which is equivalent to a cosine similarity in a transformed n-gram space. Locality sensitive hashing can be exploited for efficient retrieval of similar queries from a large query repository. This technique can be used to enhance the accuracy of query similarity calculation for rare queries, facilitate the retrieval of similar queries and significantly improve search relevance.08-09-2012
20100082507Predicting Performance Of Executing A Query In Isolation In A Database - One embodiment is a method that generates query vectors from query plans and performance vectors from data collected while executing queries in a database. The method then uses a machine learning technique (MLT) to compute distances between two query vectors and two performance vectors and to predict performance of executing a new single query in isolation in the database.04-01-2010
20100005043ACTIVE LEARNING SYSTEM, ACTIVE LEARNING METHOD AND PROGRAM FOR ACTIVE LEARNING - In order to carry out a learning in which newly acquired data is taken to be more important than data previously accumulated, a function is provided which sets a weight for learning data based on an acquisition order of the learning data. Furthermore, in order to carry out a learning which reflects data acquired in the last cycle and a result with respect to the data, a function is provided which feeds back a result of a learning in the last cycle to a rule and sets a weight for learning data based on a relation between a label of data and a prediction value.01-07-2010
20110153532DRIVING MANEUVER ASSISTING APPARATUS AND METHOD FOR ASSISTING DRIVING MANEUVER - A driving maneuver assisting apparatus includes a learning section configured to learn a driving-behavior pattern of a driver for a predetermined duration; a non-steady-state degree calculating section configured to calculate a non-steady-state degree by comparing a current driving-behavior pattern with the driving-behavior pattern learned by the learning section, wherein the non-steady-state degree represents how different the current driving-behavior pattern is from the driving-behavior pattern learned by the learning section; a learning level calculating section configured to calculate a learning level of the learning section; and a notifying section configured to notify the driver of a maneuver assisting information for inducing the driving-behavior pattern learned by the learning section in accordance with the learning level calculated by the learning level calculating section, when the non-steady-state degree calculated by the non-steady-state degree calculating section exceeds a threshold value. The notifying section is configured to provide contents of the maneuver assisting information in more detail as the learning level calculated by the learning level calculating section becomes higher.06-23-2011
20110153531INFORMATION PROCESSING APPARATUS AND CONTROL METHOD FOR THE SAME - An information processing apparatus that encodes input structured data according to an encoding rule is provided. When the structured data matches a specified learning target, this apparatus determines the start of learning of the encoding rule. Upon determining the start of learning, the apparatus recognizes the structure and data type of the structured data and starts learning the encoding rule. The apparatus stores the structured data until an end condition corresponding to the specified learning target holds and the end of learning of the encoding rule is determined. Upon determining the end of learning, the apparatus encodes the stored structured data according to the learned encoding rule.06-23-2011
20090254496SYSTEM AND METHOD FOR OPTIMIZING PATTERN RECOGNITION OF NON-GAUSSIAN PARAMETERS - A method of optimizing a function of a parameter includes associating, with an objective function for initial value of parameters, an auxiliary function of parameters that could be optimized computationally more efficiently than an original objective function, obtaining parameters that are optimum for the auxiliary function, obtaining updated parameters by taking a weighted sum of the optimum of the auxiliary function and initial model parameters.10-08-2009
20090240637RESILIENT CLASSIFIER FOR RULE-BASED SYSTEM - A resilient classifier for using with a rule-based system is provided. A system for classifying data for a rule-based system, may include: a system(s) for generating two training data sets, one data set is generated from input data while the second data set is generated from disturbed data; a system for merging the two training data sets; and a system for training a data classifier with the merged training data sets. As a result, the classification of data becomes more accurate, including when disturbed data is encountered.09-24-2009
20090187518AUTOMATICALLY IDENTIFYING AN OPTIMAL SET OF ATTRIBUTES TO FACILITATE GENERATING BEST PRACTICES FOR CONFIGURING A NETWORKED SYSTEM - A method and system for automatically identifying an optimal set of attributes of entities included in a networked system. Entity types are ranked based on information gain. A first classification accuracy relative to a first entity type is determined. The first entity type is the top-ranked entity type or a first aggregate entity type. A second entity type is selected base on the ranking. A database join of a first set of attributes associated with the first entity type and a second set of attributes associated with the second entity type is performed. A second classification accuracy relative to a second aggregate entity type generated by the join is determined. In response to determining that the second classification accuracy is not greater than the first classification accuracy, an optimal set of attributes contributing to a problem in the networked system is identified as the first set of attributes.07-23-2009
20090150309System and method for training a multi-class support vector machine to select a common subset of features for classifying objects - An improved system and method is provided for training a multi-class support vector machine to select a common subset of features for classifying objects. A multi-class support vector machine generator may be provided for learning classification functions to classify sets of objects into classes and may include a sparse support vector machine modeling engine for training a multi-class support vector machine using scaling factors by simultaneously selecting a common subset of features iteratively for all classes from sets of features representing each of the classes. An objective function using scaling factors to ensure sparsity of features may be iteratively minimized, and features may be retained and added until a small set of features stabilizes. Alternatively, a common subset of features may be found by iteratively removing at least one feature simultaneously for all classes from an active set of features initialized to represent the entire set of training features.06-11-2009
20090006286METHOD AND APPARATUS FOR IMPLEMENTING DIGITAL VIDEO MODELING TO IDENTIFY UNEXPECTED BEHAVIOR - A computer implemented method, apparatus, and computer usable program product for identifying unexpected behavioral patterns. The process parses event data derived from video data to identify behavioral patterns, wherein the event data comprises metadata describing events occurring in a selected environment. The process analyzes the behavioral patterns to identify a set of expected behavioral patterns occurring in the selected environment, and generates an expected behavioral model using the expected behavioral patterns. Thereafter, the process forms a set of unexpected behavioral patterns from the behavioral patterns inconsistent with the expected behavioral model.01-01-2009
20090119235SYSTEM AND METHOD FOR EXTRACTING ENTITIES OF INTEREST FROM TEXT USING N-GRAM MODELS - A document (or multiple documents) is analyzed to identify entities of interest within that document. This is accomplished by constructing n-gram or bi-gram models that correspond to different kinds of text entities, such as chemistry-related words and generic English words. The models can be constructed from training text selected to reflect a particular kind of text entity. The document is tokenized, and the tokens are run against the models to determine, for each token, which kind of text entity is most likely to be associated with that token. The entities of interest in the document can then be annotated accordingly.05-07-2009
20110258151System and Method for Resolving Gamma-Ray Spectra - A system for identifying radionuclide emissions is described. The system includes at least one processor for processing output signals from a radionuclide detecting device, at least one training algorithm run by the at least one processor for analyzing data derived from at least one set of known sample data from the output signals, at least one classification algorithm derived from the training algorithm for classifying unknown sample data, wherein the at least one training algorithm analyzes the at least one sample data set to derive at least one rule used by said classification algorithm for identifying at least one radionuclide emission detected by the detecting device.10-20-2011
20100306138BEHAVIOR MONITORING SYSTEM AND METHOD - A computer-implemented behavior monitoring method is provided. The method includes receiving from a plurality of contributors a plurality of personal value preference indications. Either or both of location information corresponding to a determined location of a user device and communication information corresponding to a determined communication activity of the user device are received. Either or both of the location information and the communication information of the user device are compared with the plurality of personal value preference indications from the plurality of contributors. A behavior rating is determined based on the comparison of the location information and the communication information of the user device with the plurality of personal value preference indications, and the behavior rating is transmitted to a user.12-02-2010
20080222060SYSTEM AND METHOD OF MINING TIME-CHANGING DATA STREAMS USING A DYNAMIC RULE CLASSIFIER HAVING LOW GRANULARITY - A dynamic rule classifier for mining a data stream includes at least one window for viewing data contained in the data stream and a set of rules for mining the data. Rules are added and the set of rules are updated by algorithms when an drift in a concept within the data occurs, causing unacceptable drops in classification accuracy. The dynamic rule classifier is also implemented as a method and a computer program product.09-11-2008
20080222058Bayesian-network-based method and system for detection of clinical-laboratory errors - Embodiments of the present invention include methods and systems for analyzing clinical-laboratory results and data in order to detect erroneous clinical-laboratory results. Embodiments of the present invention employ Bayesian networks and modified Bayesian networks that are constructed using cleaned clinical-laboratory results into which various types of synthetic errors have been introduced and that are optimized using different, cleaned clinical-laboratory results into which synthetic errors have been introduced.09-11-2008
20080222059COMPUTER-IMPLEMENTED METHOD, COMPUTER PROGRAM AND SYSTEM FOR ANALYZING DATA RECORDS - A computer implemented method and system for analysing a first set of data records where each data record comprises attribute values for one or more attributes, by expanding the first set of data records into a second set of data records by creating for at least one of the attributes of the first set of data records at least two redundant attributes with corresponding redundant attribute values, assigning different generalization rules to the at least two redundant attributes, and performing a generalization of the second set of data records by means of an attribute-oriented induction (AOI)-algorithm.09-11-2008
20100332423GENERALIZED ACTIVE LEARNING - Active learning is extended to decisions on information acquisition of both missing labels and missing features within one or more cases. In one example, desired (e.g., optimal) information to acquire about a case at hand and about cases in a training library during diagnostic sessions can be computed concurrently. A joint distribution of variables, comprising observed and unobserved labels and features for one or more cases, is modeled and probability distributions are determined for unobserved variables. An unobserved variable is selected from the joint distribution that has a return on information (ROI) metric having a combination of a desired uncertainty metric for a value of the unobserved variable and a desired cost for observing the value of the unobserved variable. The value of the variable is observed, and the probability distributions for the respective unobserved variables in the joint distribution are updated using the value of the identified variable.12-30-2010
20110010318SYSTEM AND METHOD FOR EMPIRICAL ENSEMBLE- BASED VIRTUAL SENSING - An empirical ensemble based virtual sensor system (VS) for the estimation of an amount of water (C) or oil (A) in a fluid mixture, said virtual sensor comprising two or more empirical models (NN01-13-2011
20110010316PROVIDING A SEAMLESS CONVERSATION SERVICE BETWEEN INTERACTING ENVIRONMENTS - An approach that provides a seamless conversation service between interacting environments is described. In one embodiment, there is a seamless conversation service tool that includes a conversation commencement component configured to facilitate commencement of a conversation between two or more parties occurring over a communication path in one of two or more interacting environments. A user context monitoring component is configured to monitor a user context associated with the conversation. A user context change identification component is configured to identify a change in the user context of the conversation. A conversation transfer component is configured to transfer the conversation between the two or more interacting environments in response to the identified change in the user context, while maintaining a transparency of functionality of the communication path.01-13-2011
20110055126Target outcome based provision of one or more templates - A computationally implemented method includes, but is not limited to: receiving one or more requests indicating at least one or more target outcomes of one or more particular templates, the one or more particular templates designed to facilitate one or more end users to achieve the one or more target outcomes when one or more emulatable aspects included in the one or more particular templates are emulated; and providing from a plurality of templates the one or more particular templates, the providing being based at least on the one or more particular templates' association with the one or more target outcomes, the one or more particular templates developed based on one or more reported aspects of one or more source users In addition to the foregoing, other method aspects are described in the claims, drawings, and text forming a part of the present disclosure.03-03-2011
20110055125Template development based on sensor originated reported aspects - A computationally implemented method includes, but is not limited to: providing one or more reported aspects associated with one or more source users that were originally reported by one or more sensors; and developing one or more templates designed to facilitate one or more end users to achieve one or more target outcomes when one or more emulatable aspects indicated by the one or more templates are emulated, the development of the one or more templates being based at least on a portion of the one or more reported aspects In addition to the foregoing, other method aspects are described in the claims, drawings, and text forming a part of the present disclosure.03-03-2011
20110055121SYSTEM AND METHOD FOR IDENTIFYING AN OBSERVED PHENEMENON - A system for identifying an observed phenomenon. The system includes a computing device configured for receiving disparate data streams associated with disparate data sources. The system also includes a feature extraction module communicably connected to the computing device, a classification module communicably connected to the computing device, and a consensus module communicably connected to the computing device. The feature extraction module is configured for generating a set of attributes for each data stream. The classification module is configured for soft associating labels with attributes for each set of attributes, and for generating a confidence value for each soft association. The consensus module is configured for generating an output indicative of the phenomenon. The consensus module includes a standardization module and a sequential data module. The standardization module is configured for standardizing the confidence values. The sequential data module is configured for generating the output based on the standardized confidence values.03-03-2011
20110055124Development of personalized plans based on acquisition of relevant reported aspects - A computationally implemented method includes, but is not limited to: acquiring one or more relevant reported aspects associated with one or more source users that are relevant to achieving one or more target outcomes, the acquisition of the one or more relevant reported aspects being based, at least in part, on relevancy of the one or more relevant reported aspects with respect to the achievement of the one or more target outcomes; and developing one or more personalized plans designed to facilitate an end user to achieve the one or more target outcomes when one or more emulatable aspects indicated by the one or more personalized plans are emulated, the development of the one or more personalized plans being based, at least in part, on the acquiring In addition to the foregoing, other method aspects are described in the claims, drawings, and text forming a part of the present disclosure.03-03-2011
20110055123Systems and Methods for Using Multiple In-line Heuristics to Reduce False Positives - An exemplary method for using multiple in-line heuristics to reduce false positives may include: 1) training a first heuristic using a set of training data, 2) deploying the first heuristic, 3) identifying false positives produced by the first heuristic during deployment, 4) modifying the training data to include the false positives produced by the first heuristic, 5) creating a second heuristic using the modified training data, 6) deploying both the first heuristic and the second heuristic, and then 7) applying both the first heuristic and the second heuristic, in sequence, to a set of field data.03-03-2011
20110016066AUTOMATIC CONFIGURATION AND CONTROL OF DEVICES USING METADATA - Particular embodiments generally relate to automatically controlling an item. For example, items may include electronic devices, such as televisions, lights, etc, and/or virtual devices, such as applications, etc. In one embodiment, items may be configured using metatags. When a device is connected for operation, one or more metatags for the device are received. A metatag be used to classify the device. For example, the metatag may indicate uses, locations, connections, etc. The use of device (e.g., pathway, reading, etc.) classifies the item in way it can be used. For example, a user may use a pathway light in different ways, such as the user may turn on all lights with the pathway metatag at night. The location indicates the location of the item, such as in the living room, bedroom, etc. The connections indications a type of item, such as a bedroom light, lamp, TV, etc.01-20-2011
20110022553DIAGNOSIS SUPPORT SYSTEM, DIAGNOSIS SUPPORT METHOD THEREFOR, AND INFORMATION PROCESSING APPARATUS - A diagnosis support system includes a learning unit which calculates a first learning result based on diagnosis results on case data which are obtained by a plurality of doctors and a second learning result based on a diagnosis result on the case data which is obtained by a specific doctor, an analysis unit which analyzes a feature associated with diagnosis by the specific doctor based on a comparison between the first learning result and the second learning result, and a decision unit which decides display information of clinical data obtained by examination of a patient based on the analysis result.01-27-2011
20110022550MIXING KNOWLEDGE SOURCES WITH AUTO LEARNING FOR IMPROVED ENTITY EXTRACTION - The disclosed embodiments of computer systems and techniques utilize an ensemble semantics framework to combine knowledge acquisition systems that yield significantly higher quality resources than each system in isolation. Gains in entity extraction are achieved by combining state-of-the-art distributional and pattern-based systems with a large set of features from, for example, a webcrawl, query logs, and wisdom of the crowd sources. This results in improved query interpretation and greater relevancy in providing search results and advertising, for example.01-27-2011
20110022549Presenting Search Results Based on User-Customizable Criteria - In one embodiment, ranking search results generated in response to search queries comprises: receiving, a search query from a user; identifying a plurality of network contents in response to the search query; determining one or more ranking criteria for the search query; presenting the ranking criteria to the user; receiving from the user one or more weights assigned to one or more of the ranking criteria; ranking the identified network contents based on the ranking criteria and the weights; and presenting the network contents to the user in an order according to their ranking.01-27-2011
20110258148ACTIVE PREDICTION OF DIVERSE SEARCH INTENT BASED UPON USER BROWSING BEHAVIOR - Many search engines attempt to understand and predict a user's search intent after the submission of search queries. Predicting search intent allows search engines to tailor search results to particular information needs of the user. Unfortunately, current techniques passively predict search intent after a query is submitted. Accordingly, one or more systems and/or techniques for actively predicting search intent from user browsing behavior data are disclosed herein. For example, search patterns of a user browsing a web page and shortly thereafter performing a query may be extracted from user browsing behavior. Queries within the search patterns may be ranked based upon a search trigger likelihood that content of the web page motivated the user to perform the query. In this way, query suggestions having a high search trigger likelihood and a diverse range of topics may be generated and/or presented to users of the web page.10-20-2011
20110264609PROBABILISTIC GRADIENT BOOSTED MACHINES - Probabilistic gradient boosted machines are described herein. A probabilistic gradient boosted machine can be utilized to learn a function based at least in part upon sets of observations of a target attribute that is common across a plurality of entities and feature vectors that are representative of such entities. The sets of observations are assumed to accord to a distribution function in the exponential family. The learned function is utilized to generate values that are employed parameterize the distribution function, such that sets of observations can be predicted for different entities.10-27-2011
20110078099METHOD FOR FEATURE SELECTION AND FOR EVALUATING FEATURES IDENTIFIED AS SIGNIFICANT FOR CLASSIFYING DATA - A group of features that has been identified as “significant” in being able to separate data into classes is evaluated using a support vector machine which separates the dataset into classes one feature at a time. After separation, an extremal margin value is assigned to each feature based on the distance between the lowest feature value in the first class and the highest feature value in the second class. Separately, extremal margin values are calculated for a normal distribution within a large number of randomly drawn example sets for the two classes to determine the number of examples within the normal distribution that would have a specified extremal margin value. Using p-values calculated for the normal distribution, a desired p-value is selected. The specified extremal margin value corresponding to the selected p-value is compared to the calculated extremal margin values for the group of features. The features in the group that have a calculated extremal margin value less than the specified margin value are labeled as falsely significant.03-31-2011
20100293117METHOD AND SYSTEM FOR FACILITATING BATCH MODE ACTIVE LEARNING - A method and system for performing batch mode active learning to train a classifier. According to embodiments of the present invention, unlabeled documents are selected from a corpus based on rewards associated with each unlabeled document. The reward is an indication of the increase to the accuracy of a classifier which may result if the document is used to train the classifier. When calculating a given reward, embodiments of the present invention address the uncertainty and diversity of a given document. Embodiments of the present invention reduce the resources utilized to perform classifier training.11-18-2010
20100293118METHODS AND SYSTEMS FOR PREDICTING PROTEIN-LIGAND COUPLING SPECIFICITIES - The invention provides methods and systems for predicting or evaluating protein-ligand coupling specificities. A pattern recognition model can be trained by selected sequence segments of training proteins which have a specified ligand coupling specificity. Each selected sequence segment is believed to include amino acid residue(s) that may contribute to the ligand coupling specificity of the corresponding training protein. Sequence segments in a protein of interest can be similarly selected and used to query the trained model to determine if the protein of interest has the same ligand coupling specificity as the training proteins. In one embodiment, the pattern recognition model employed is a hidden Markov model which is trained by concatenated cytosolic domains of GPCRs which have interaction preference to a specified class of G proteins. This trained model can be used to evaluate G protein coupling specificity of orphan GPCRs.11-18-2010
20100293115METHOD, SYSTEM AND APPARATUS FOR REAL-TIME CLASSIFICATION OF MUSCLE SIGNALS FROM SELF -SELECTED INTENTIONAL MOVEMENTS - A new method, system and apparatus is provided that enables muscle signals that correspond to muscle contractions to be mapped to one or more functions of an electronic device such as a prosthetic device or gaming apparatus. Muscle signals are classified in real-time from self-selected intentional movements. A self-training protocol allows users to select and label their own muscle contractions, and is operable to automatically determine the discernible and repeatable muscle signals generated by the user. A visual display means is used to provide visual feedback to users illustrating the responsiveness of the system to muscle signals generated by the user.11-18-2010
20110119210Multiple Category Learning for Training Classifiers - Described is multiple category learning to jointly train a plurality of classifiers in an iterative manner. Each training iteration associates an adaptive label with each training example, in which during the iterations, the adaptive label of any example is able to be changed by the subsequent reclassification. In this manner, any mislabeled training example is corrected by the classifiers during training. The training may use a probabilistic multiple category boosting algorithm that maintains probability data provided by the classifiers, or a winner-take-all multiple category boosting algorithm selects the adaptive label based upon the highest probability classification. The multiple category boosting training system may be coupled to a multiple instance learning mechanism to obtain the training examples. The trained classifiers may be used as weak classifiers that provide a label used to select a deep classifier for further classification, e.g., to provide a multi-view object detector.05-19-2011
20110119209METHOD AND SYSTEM FOR DEVELOPING A CLASSIFICATION TOOL - An exemplary embodiment of the present invention provides a computer implemented method of developing a classifier. The method includes obtaining a set of training data comprising labeled cases. The method also includes training a classifier based, at least in part, on the training data. The method also includes applying the classifier to a plurality of unlabeled cases to generate classification scores for each of the unlabeled cases, wherein each classification score corresponds with an instance of a corresponding case. Furthermore, the classification score corresponding to a first instance in a case is computed based, at least in part, on a value of a case-centric feature corresponding to the first instance, wherein the value of the case-centric feature is based, at least in part, on characteristics of the first instance and a second instance in the case.05-19-2011
20110119211SYSTEM AND METHOD FOR ASSESSING RISK - The invention describes systems and methods of assessing risk using a computer. A computer-based system including an enrollment module, a data aggregation module, a risk assessment module, and a memory is provided for assessing risks. The enrollment receives, at a computer, personal information regarding at least one entity. The data aggregation module receives, at the computer, risk information regarding the entity according to the personal information from at least one data source. The risk assessment module converts the risk information to assessment information. The memory stores the personal information, the risk information, and/or the assessment information on the computer.05-19-2011
20120150773USER INTERFACE AND WORKFLOW FOR PERFORMING MACHINE LEARNING - A computing device receives a training data set that includes a plurality of positive examples of sensitive data and a plurality of negative examples of sensitive data via a user interface. The computing device analyzes the training data set using machine learning to generate a machine learning-based detection (MLD) profile that can be used to classify new data as sensitive data or as non-sensitive data. The computing device displays a quality metric for the MLD profile in the user interface.06-14-2012
20120278263COST-SENSITIVE ALTERNATING DECISION TREES FOR RECORD LINKAGE - Record Linkage (RL) is the task of identifying two or more records referring to the same entity (e.g., a person, a company, etc.). RL models can be based on Cost Sensitive Alternating Decision Trees (ADTree), an algorithm that uniquely combines boosting and decision trees algorithms to create shorter and easier-to-interpret linking rules. These models can be naturally trained to operate at industrial precision/recall operating points, and the shorter output rules are so clear that it can effectively explain its decisions to non-technical users via score aggregation or visualization. The models significantly outperform other baselines on the desired industrial operating points, and the improved understanding of the model's decisions led to faster debugging and feature development cycles.11-01-2012
20120278264TECHNIQUES TO FILTER MEDIA CONTENT BASED ON ENTITY REPUTATION - Techniques to filter media content based on entity reputation are described. An apparatus may comprise a reputation subsystem that manages an entity reputation score for an entity. The reputation subsystem comprising a reputation manager component and a reputation input/output (I/O) component. The reputation manager component may include a data collection module that collects reputation information for an entity from a selected set of multiple reputation sources. The reputation manager component may also comprise a feature manager module, communicatively coupled to the data collection module, that extracts a selected set of reputation features from the reputation information. The reputation manager component may further comprise a reputation scoring module, communicatively coupled to the feature manager module, that generates an entity reputation score based on the reputation features using a supervised or unsupervised machine learning algorithm. Other embodiments are described and claimed.11-01-2012
20110125682LEARNING DEVICE - Provided is a learning device that represents the learning contents of a learning edition in visual and audio elements such as voice, melody, and image. The learning contents are pointed out by an indicator according to the principle of signal transmission and reception for recognizing the position of the indicator.05-26-2011
20100262571SYSTEMS AND METHODS FOR ORGANIZING DATA SETS - A method is provided for organizing data sets. In use, an automatic decision system is created or updated for determining whether data elements fit a predefined organization or not, where the decision system is based on a set of preorganized data elements. A plurality of data elements is organized using the decision system. At least one organized data element is selected for output to a user based on a score or confidence from the decision system for the at least one organized data element. Additionally, at least a portion of the at least one organized data element is output to the user. A response is received from the user comprising at least one of a confirmation, modification, and a negation of the organization of the at least one organized data element. The automatic decision system is recreated or updated based on the user response. Other embodiments are also presented.10-14-2010
20100262569DATA CONVERTING APPARATUS AND MEDIUM HAVING DATA CONVERTING PROGRAM - A data converting apparatus and a data converting program are suitable to quantitatively estimate, based on a temporal alteration of a stimulus value, a temporal alteration of a sensitivity brought to human beings. The data converting apparatus includes a decomposing unit subjecting temporal alteration data of a stimulus value to wavelet decomposition to extract plural time-frequency components contained in the temporal alteration data, a weighting unit weighting the plural extracted time-frequency components weighting coefficients which are predetermined based on a relationship between a temporal alteration of a stimulus value and a temporal alteration of a sensitivity of human beings to the stimulus value, and a synthesizing unit subjecting the plural weighted time-frequency components to wavelet synthesis to estimate a sensitivity brought to human beings when the stimulus value is subjected to temporal alteration according to the temporal alteration data.10-14-2010
20100262568Scalable Clustering - A scalable clustering system is described. In an embodiment the clustering system is operable for extremely large scale applications where millions of items having tens of millions of features are clustered. In an embodiment the clustering system uses a probabilistic cluster model which models uncertainty in the data set where the data set may be for example, advertisements which are subscribed to keywords, text documents containing text keywords, images having associated features or other items. In an embodiment the clustering system is used to generate additional features for associating with a given item. For example, additional keywords are suggested which an advertiser may like to subscribe to. The additional features that are generated have associated probability values which may be used to rank those features in some embodiments. User feedback about the generated features is received and used to revise the feature generation process in some examples.10-14-2010
20100185571INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing apparatus includes a plurality of information input units inputting information including image information or sound information in a real space, an event detection unit analyzing input information from the information input units so as to generate event information including estimated position information and estimated identification information of users present in the real space, and an information integration processing unit setting hypothesis data regarding user existence and position information and user identification information of the users in the real space and updating and selecting hypothesis data based on the event information so as to generate analysis information including user existence and position information and user identification information of the users in the real space.07-22-2010
20100185568Method and System for Document Classification - A system and method to classify web-based documents as articles or non-articles is disclosed. The method generates a machine learning model from a human labelled training set which contains articles and non-articles. The machine learning model is applied to new articles to label them as articles or non-articles. The method generates the machine learning model based on content, such as text and tags of the web-based documents. The invention also provides for devices which incorporate the machine learning model, allowing such devices to classify documents as articles or non-articles.07-22-2010
20100179929SYSTEM FOR FINDING QUERIES AIMING AT TAIL URLs - Systems and methodologies for improved query classification and processing are provided herein. As described herein, a query prediction model can be constructed from a set of training data (e.g., diagnostic data obtained from an automatic diagnostic system and/or other suitable data) using a machine learning-based technique. Subsequently upon receiving a query, a set of features corresponding to the query, such as the length and/or frequency of the query, unigram probabilities of respective words and/or groups of words in the query, presence of pre-designated words or phrases in the query, or the like, can be generated. The generated features can then be analyzed in combination with the query prediction model to classify the query by predicting whether the query is aimed at a head Uniform Resource Locator (URL) or a tail URL. Based on this prediction, an appropriate index or combination of indexes can be assigned to answer the query.07-15-2010
20120209794SELF-ORGANIZING SEQUENTIAL MEMORY PATTERN MACHINE AND REINFORCEMENT LEARNING METHOD - A self-organizing computing machine utilizes a method for mapping from a plurality of patterns contained within provided inputs to an invariant perception, distinguishable by a name or a label. The self-organizing computing machine includes a network of at least three nodes arranged in at least two hierarchical levels, at least one feature extractor, and at least one output unit arranged to interface the invariant perception. The nodes may include a reinforcement learning sub-network combined with an ensemble learning sub-network. The reinforcement learning sub-network may be arranged to receive at least two correlants, to determine a plurality of output values and to output the output values to the nodes of the higher level and the nodes of the lower level. Also, the ensemble learning sub-network may be arranged to receive and to combine output values from nodes of the higher level and nodes of the lower level.08-16-2012
20120310869ACTIVE LEARNING SYSTEMS AND METHODS FOR RAPID PORTING OF MACHINE TRANSLATION SYSTEMS TO NEW LANGUAGE PAIRS OR NEW DOMAINS - Systems and methods for active learning of statistical machine translation systems through dynamic creation and updating of parallel corpus. The systems and methods provided create accurate parallel corpus entries from a test set of sentences, words, phrases, etc. by calculating confidence scores for particular translations. Translations with high confidence scores are added directly to the corpus and the translations with low confidence scores are presented to human translations for corrections.12-06-2012
20120310868METHOD AND SYSTEM FOR EXTRACTING AND MANAGING INFORMATION CONTAINED IN ELECTRONIC DOCUMENTS - A method and system that utilize metadata to facilitate extraction and enable management of information contained in electronic documents. Metadata describe content of documents based on composition of their structure and ways information is arranged in a structure. The system makes it possible to automatically manage models used for extraction, and metadata also define a logical schema for managing information extracted. The method includes a preparation step in which metadata and document samples are collected and stored, followed by a training step in which the system utilizes metadata and respective document samples to build and train models used for extraction. Finally, in an extraction step, the system receives a collection of documents and utilizes trained models to extract information that can be stored according to logical schema defined from metadata and can be immediately managed. The system enables methods to be applied to information dispersed throughout large documents. In one preferred embodiment, metadata is supplied by an XSD (XML Schema Definition) and document samples are labeled in a XML format that can be validated by the XSD.12-06-2012
20120310867Method for Learning Remote Control and Learning Remote Control Thereof - A learning method and a learning remote control in the present invention include: selecting keys to be taught on the learning remote control, and constructing a key series; performing such operation to each key to be taught as: prompting to the keys to be taught, and then receiving and learning commands from a teaching remote control according to the key to be taught; and when being detected that a key is further selected during the operation to each key to be taught, renewing the key series according to this key, and reexecute the operation to each key. The user follows flexible and simple operating steps during the learning procedure of the learning remote control. The operation is brief, and the learning procedure is simple. The intelligent performance of the learning remote control is improved.12-06-2012
20120310866DATA PROCESSING DEVICE, COMPUTER PROGRAM THEREFOR AND DATA PROCESSING METHOD - A plurality of pruning measures (PM) are calculated from a feature amount (CV) of test data (TD) which is input, a plurality of isopycnic surfaces (EC) are plotted and set on a threshold space (SS), a threshold curved surface (SC) in which a decrease in at least one of a plurality of pruning measures (PM) causes an increase in at least one thereof is generated using a portion of one isopycnic surface (EC) as a part, a hypothesis curved surface (HC) of subject data (CD) is generated on the threshold space (SS) to set a position intersecting the threshold curved surface (SC) to a pruning threshold (PS), and a plurality of hypotheses of the subject data (CD) are pruned. Thereby, there is provided a data processing device of which at least one of the recognition speed and the recognition accuracy is higher than in the related art.12-06-2012
20120310865METHOD FOR GENERATING A MACHINE HEARTBEAT - A method and system for generating a heartbeat of a process including at least one machine configured to perform a process cycle consisting of a plurality of timed events performed in a process sequence under an identified condition includes determining the duration of each of the timed events during the process cycle performed under the identified condition, ordering the durations of the plurality of timed events in the process sequence, and generating a heartbeat defined by the ordered durations of a process cycle. The identified condition may be one of a design intent, baseline, learnt, known, current or prior condition. The variance of the heartbeat between a first and at least a second identified condition may be analyzed to monitor and/or control the process or machine. The system may display the process heartbeat information and may generate a message in response to the heartbeat and/or variance thereof.12-06-2012
20120310864Adaptive Batch Mode Active Learning for Evolving a Classifier - This disclosure includes various embodiments of apparatuses, systems, and methods for adaptive batch mode active learning for evolving a classifier. A corpus of unlabeled data elements to be classified is received, a batch size is determined based on a score function, a batch of unlabeled data elements having the determined batch size is selected from the corpus and labeled using a labeling agent or oracle, a classifier is retrained with the labeled data elements, these steps are repeated until a stop criterion has been met, for example, the classifier obtains a desired performance on unlabeled data elements in the corpus. The batch size determination and selection of a batch unlabeled data elements may be based on a single score function. The data elements may be video, image, audio, web text, and/or other data elements.12-06-2012
20120310863GENE-SPECIFIC PREDICTION - A gene-specific prediction tool for classifying and interpreting gene tests is described. The prediction tool includes a classifier trained and tested using databases of gene variants and their known phenotypes. The classifier uses differences between features of amino acids in obtaining attributes used to perform classification and generate predictions, including for benign and pathologic outcomes, for uncertain gene variants.12-06-2012
20100023464Systems and methods for parameter adaptation - A method of parameter adaptation is used to modify the parameters of a model to improve model performance. The model separately estimates the contribution of each model parameter to the prediction error. It achieves this by transforming to the time-scale plane the vectors of output sensitivities with respect to model parameters and then identifying the regions within the time-scale plane at which the dynamic effect of individual model parameters is dominant on the output. The method then attributes the prediction error in these regions to the deviation of a single parameter from its true value as the basis of estimating individual parametric errors. The proposed Signature Isolation Method (PARSIM) then uses these estimates to adapt individual model parameters independently of the others, implementing, in effect, concurrent adaptation of individual parameters by the Newton-Raphson method in the time-scale plane. The Signature Isolation Method has been found to have an adaptation precision comparable to that of the Gauss-Newton method for noise-free cases. However, it surpasses the Gauss-Newton method in precision when the output is contaminated with noise or when the measurements are generated by inadequate excitation.01-28-2010
20110137833DATA PROCESSING APPARATUS, DATA PROCESSING METHOD AND PROGRAM - The data processing apparatus includes a state series generation unit and a computing unit. The state series generation unit generates a time series data of state nodes from a time series data of event. The state transition model of the event is expressed as a stochastic state transition model. The computing unit computes the parameters for the stochastic state transition model of events by computing parameters of time series data corresponding to an appearance frequency of the state nodes, the appearance frequency of transitions among the state nodes and the like.06-09-2011
20090089228GENERALIZED REDUCED ERROR LOGISTIC REGRESSION METHOD - A machine classification learning method titled Generalized Reduced Error Logistic Regression (RELR) is presented. The method overcomes significant limitations in prior art logistic regression and other machine classification learning methods. The method is applicable to all current applications of logistic regression, but has significantly greater accuracy using smaller sample sizes and larger numbers of input variables than other machine classification learning methods including prior art logistic regression.04-02-2009
20090204557Method and System for Analysis of Flow Cytometry Data Using Support Vector Machines - An automated method and system are provided for receiving an input of flow cytometry data and analyzing the data using one or more support vector machines to generate an output in which the flow cytometry data is classified into two or more categories. The one or more support vector machines utilizes a kernel that captures distributional data within the input data. Such a distributional kernel is constructed by using a distance function (divergence) between two distributions. In the preferred embodiment, a kernel based upon the Bhattacharya affinity is used. The distributional kernel is applied to classification of flow cytometry data obtained from patients suspected having myelodysplastic syndrome.08-13-2009
20110137836METHOD AND SYSTEM FOR GENERATING HISTORY OF BEHAVIOR - Disclosed are method and system for generating history of behavior that is capable of simplifying input of a behavior content of a human behavior pattern determined from data measured by a sensor. A computer obtains biological information measured by a sensor which is mounted to a person and accumulates the biological information, obtains motion frequencies from the accumulated biological information, obtains time-series change points of the motion frequencies, extracts a period between the change points as a scene which is a period of being in the state of an identical motion, compares the motion frequencies with a preset condition for each extracted scene and identifies the action contents in the scene, estimates the behavior content performed by the person in the scene on the basis of the appearance sequence of the action contents, and generates the history of the behaviors on the basis of the estimated behavior contents.06-09-2011
20110137835INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing device includes an acquisition unit acquiring a viewing log including information representing content of an operation for viewing content and time of the operation, a learning unit learning, based on the viewing log acquired by the acquisition unit, a viewing behavior model which is a stochastic state transition model representing a viewing behavior of a user, a recognition unit recognizing, using the viewing behavior model obtained through learning by the learning unit, a current viewing state of the user, a prediction unit predicting, using the viewing behavior model, the viewing behavior of the user after a predetermined period of time with the current viewing state of the user recognized by the recognition unit as a starting point, and a display control unit displaying information relating to content predicted to be viewed through the viewing behavior predicted by the prediction unit.06-09-2011
20110137834LEARNING APPARATUS AND METHOD, PREDICTION APPARATUS AND METHOD, AND PROGRAM - A learning apparatus includes: a location acquiring section for acquiring time series data on locations of a user; a time acquiring section for acquiring time series data on times; and learning section for learning an activity model indicating an activity state of the user as a probabilistic state transition model, using the respective acquired time series data on the locations and the times as an input.06-09-2011
20110137832RELATIONAL BAYESIAN MODELING FOR ELECTRONIC COMMERCE - The present invention provides a language, method and system to formulate and evaluate relational Bayesian networks in an e-commerce environment. The present invention employs a specific language for constructing synthetic variables used to predict events in the Bayesian networks. The present system and language allow for efficient and accurate representation, inference, and discovery of the synthetic variables used to model web visitor behavior.06-09-2011
20110137831LEARNING APPARATUS, LEARNING METHOD AND PROGRAM - A learning apparatus includes: an interpolating section which interpolates data missing in time series data; an estimating section which estimates a Hidden Markov Model from the time series data; and a likelihood calculating section which calculates the likelihood of the estimated Hidden Markov Model. The likelihood calculating section calculates the likelihood for normal data which does not have missing data and the likelihood for interpolation data which is interpolated data in different conditions and calculates the likelihood of the Hidden Markov Model for the time series data in which the data is interpolated. The estimating section updates the Hidden Markov Model so that the likelihood calculated by the likelihood calculating section becomes high.06-09-2011
20110137830FRAMEWORK FOR FINDING ONE OR MORE SOLUTIONS TO A PROBLEM - In an embodiment, information for use in identifying a plurality of sub-solvers may be acquired. The plurality of sub-solvers may be used in a first attempt to find at least one solution to a problem that may be defined in the acquired information. At least two of the sub-solvers in the plurality of sub-solvers may be of different sub-solver types. The sub-solvers may be identified based on the acquired information. One or more starting points for the identified sub-solvers may be identified and transferred to the identified sub-solvers. One or more outputs, that indicate one or more results associated with the first attempt to find at least one solution to the problem, may be acquired from the identified sub-solvers. One or more sub-solvers may be identified, based on the acquired one or more outputs, for use in a second attempt to find at least one solution to the problem.06-09-2011
20110178966Method for Matching Elements in Schemas of Databases Using a Bayesian Network - A method matches elements in two schemas for two associated databases using automatic schema matching (ASM), wherein there is one schema for each database, wherein the elements define objects in the databases, and wherein the matching is performed on pairs of the elements by a combined matcher including a set of matchers. A Bayesian network (BN) is constructed for the set of matchers, and for each pair of elements the following steps are performing: obtaining an individual similarity value for each pair of the elements and each matcher, determining a likelihood ratio for each individual similarity value, performing belief updating on the BN using the likelihood ratios to obtain a final similarity value and corresponding probability, and outputting the final similarity value and the probability to indicate whether the pair of the elements match, or not.07-21-2011
20110178965METHOD FOR TRAINING A SYSTEM TO SPECIFICALLY REACT ON A SPECIFIC INPUT - A method for training a system to specifically react on a specific input. The method can include defining a set of binary data structures, each representing a real-world component, item, or virtual object; storing each data structure as a binary pattern; creating uniquely identifiable copies of the data structures to represent individual instances of the components, items, or virtual objects; creating a virtual state space of the components, items, or virtual objects by grouping them as relevant for a specific situation; receiving an input to change a status or an attribute value of at least one of the components, items, or virtual objects; storing the received changes in a new version of the applicable data structure instance; analyzing similarities of the stored binary patterns related to a particular action performed; and if a matched binary pattern is identified, proposing at least one possible action related to the matched binary pattern.07-21-2011
20110178963 SYSTEM FOR THE DETECTION OF RARE DATA SITUATIONS IN PROCESSES - An apparatus for detecting a rare situation in a process described by a plurality of parameters, the apparatus comprising: a parameter value inputter, for inputting values of at least two interrelated parameters of the plurality of parameters, the interrelated parameters constituting at least one cluster, and a rare situation detector for detecting a rare situation according to an alert policy, the alert policy being based at least on an output value of an alert model, the alert model configured to provide the output value as a function of the input parameter values of parameters constituting the at least one cluster.07-21-2011
20100161527EFFICIENTLY BUILDING COMPACT MODELS FOR LARGE TAXONOMY TEXT CLASSIFICATION - A taxonomy model is determined with a reduced number of weights. For example, the taxonomy model is a tangible representation of a hierarchy of nodes that represents a hierarchy of classes that, when labeled with a representation of a combination of weights, is usable to classify documents having known features but unknown class. For each node of the taxonomy, the training example documents are processed to determine the features for which there are a sufficient number of training example documents having a class label corresponding to at least one of the leaf nodes of a subtree having that node as a root node. For each node of the taxonomy, a sparse weight vector is determined for that node, including setting zero weights, for that node, those features determined to not appear at least a minimum number of times in a given set of leaf nodes in the sub-tree with that node as a root node. The sparse weight vectors can be learned by solving an optimization problem using a maximum entropy classifier, or a large margin classifier with a sequential dual method (SDM) with margin or slack resealing. The determined sparse weight vectors are tangibly embodied in a computer-readable medium in association with the tangible representation of the nodes of the taxonomy.06-24-2010
20100161529Self-Calibration - Mitigation of processing artefacts caused by surfaces with high contrast printing or colouring transitions within a system to compare signatures derived from inherent physical surface properties of different articles to authenticate or validate articles and within a system to generate signatures from inherent physical surface properties of different articles.06-24-2010
20100161524METHOD AND SYSTEM FOR IDENTIFYING GRAPHICAL MODEL SEMANTICS - A system and method for identifying graphical model semantics, one aspect, receive a graphical diagram, associate each of a plurality of elements with one or more predetermined meta-types, identify a plurality of types in the graphical diagram, and determine a category for each of elements in said graphical diagram. Containment identification rules identify one or more containment relationships in the graphical diagram. Multiplicity identification rules identify multiplicity relationships in the graphical diagram. Advanced semantic rules identify visual elements that represent attributes and refine relationships to identify unique behavior.06-24-2010
20100121793PATTERN GENERATION METHOD, PATTERN GENERATION APPARATUS, AND PROGRAM - Disclosed is an apparatus that generates automatically a characteristic pattern in time series data by clustering a plurality of time series subsequences generated from the time series data. The apparatus includes a time series subsequence generation unit that generates a plurality of time series subsequences from the time series data, a phase alignment unit that aligns a phase of the generated time series subsequence, a clustering unit that performs clustering of a plurality of the time series subsequences, each having a phase aligned, a storage apparatus that stores the pattern obtained by the clustering, and an output apparatus that outputs the stored pattern.05-13-2010
20090299925Automatic Detection of Undesirable Users of an Online Communication Resource Based on Content Analytics - An exemplary processor-implemented method of determining whether a user of an online communication resource is an undesirable user includes the steps of building at least one model based on at least one feature of a feature set using at least one machine learning technique; and classifying the user by comparing at least one feature of the feature set that is associated with the user to the at least one model, a determination as to whether the user is an undesirable user being based at least in part on the classification of the user.12-03-2009
20090299924INTELLIGENT HUMAN-MACHINE INTERFACE - Embodiments in accordance with the present invention relate to methods and apparatus for an intelligent human-machine interface. By way of example, but not limited thereto, embodiments of methods and apparatus are presented of an intelligent human-machine interface for the operating room, and more particularly, to systems and processes for real-time management and feedback of process control, situational awareness, logistics, communication, and documentation, herein referred to as system. One element of the system, among others, provides a knowledge base that organizes information and rules that enables an accurate, relevant and timely decision support system. The knowledge base is represented in a hierarchical structure of functions and systems. The system serves as platform for the avoidance, detection and timely correction of errors, and as such, acts as a countermeasure to error.12-03-2009
20090292660USING RULE INDUCTION TO IDENTIFY EMERGING TRENDS IN UNSTRUCTURED TEXT STREAMS - A method for identifying emerging concepts in unstructured text streams comprises: selecting a subset V of documents from a set U of documents; generating at least one Boolean combination of terms that partitions the set U into a plurality of categories that represent a generalized, statistically based model of the selected subset V wherein the categories are disjoint inasmuch as each document of U is included in only one category of the partition; and generating a descriptive label for each of the disjoint categories from the Boolean combination of terms for that category.11-26-2009
20110137829Method for Selecting Neighborhoods of Training Points for Local Learning - A method selects a subset of training points near a query point from a set of training points. The subset of training points near the query point is determined from a the set of training points such that a cumulative similarity is maximized, wherein the cumulative similarity measures a similarity of the query point to each point in the subset and a similarity of points in the subset to each other.06-09-2011
20100030715Social Network Model for Semantic Processing - A social network model, based on data relevant to a user, is used for semantic processing to enable improved entity recognition among text accessed by the user. An entity extraction module of the server, with reference to a general training corpus, general gazetteers, user-specific gazetteers, and entity models, parses text to identify entities. The entities may be, for example, people, organizations, or locations. A social network module of the server builds the social network model implicit in the data accessed by the user. The social network model includes the relationships between entities and an indication of the strength of each relationship. The social network module is also used to disambiguate names and unify entities based on the social network model.02-04-2010
20100030714METHOD AND SYSTEM TO IMPROVE AUTOMATED EMOTIONAL RECOGNITION - An automated emotional recognition system includes an emotional state classifier adapted to receive, during an operative phase, an input information stream with embedded information related to emotional states of a person, and to generate a succession of emotional state indications derived from the input information stream. The emotional recognition system further includes a post-processing function, configured to receive at least two emotional state indications of the succession and, for each of said at least two emotional state indications, determine a corresponding emotional state representation in an emotional state representation system. The post-processing function is further configured to combine the emotional state representations of the at least two emotional state indications to obtain an output emotional state indication.02-04-2010
20100023466RULE LEARNING METHOD, PROGRAM AND APPARATUS - A rule learning method for making a computer perform rule learning processing in machine learning includes firstly calculating an evaluation value of respective features in a training example by using data and weights of the training examples; selecting a given number of features in descending order of the evaluation values; secondly calculating a confidence value for one of the given number of selected features; updating the weights of training example, by using the data and weights of the training examples, and the confidence value corresponding to the one feature; firstly repeating the updating for the remaining features of the given number of features; and secondly repeating, for a given number of times, the firstly calculating, the selecting, the secondly calculating, the updating, and the firstly repeating.01-28-2010
20100017350Method and Apparatus for Automatically Structuring Free Form Heterogeneous Data - Techniques are provided for automatically structuring free form heterogeneous data. In one aspect of the invention, the techniques include obtaining free form heterogeneous data, segmenting the free form heterogeneous data into one or more units, automatically labeling the one or more units based on one or more machine learning techniques, wherein each unit is associated with a label indicating an information type, and structuring the one or more labeled units in a format to facilitate one or more operations that use at least a portion of the labeled units, e.g., information technology (IT) operations.01-21-2010
20100145894GENOMIC CLASSIFICATION OF COLORECTAL CANCER BASED ON PATTERNS OF GENE COPY NUMBER ALTERATIONS - The invention is directed to methods and kits that allow for classification of colorectal cancer cells according to genomic profiles, and methods of diagnosing, predicting clinical outcomes, and stratifying patient populations for clinical testing and treatment using the same.06-10-2010
20080243728Recursive Feature Eliminating Method Based on a Support Vector Machine - Method, apparatus and system are described to perform a feature eliminating method based on a support vector machine. In some embodiments, a value for each feature in a group of features provided by a training data is determined. At least one feature is eliminated from the group by utilizing the value for each feature in the group. The value for each feature in the group is updated based upon a part of the training data that corresponds to the eliminated feature.10-02-2008
20110307427Molecular markers predicting response to adjuvant therapy, or disease progression, inbreast cancer - Predicting response to adjuvant therapy or predicting disease progression in breast cancer is realized by (1) first obtaining a breast cancer test sample from a subject; (2) second obtaining clinicopathological data from said breast cancer test sample; (3) analyzing the obtained breast cancer test sample for presence or amount of (a) one or more molecular markers of hormone receptor status, one or more growth factor receptor markers, (b) one or more tumor suppression/apoptosis molecular markers; and (c) one or more additional molecular markers both proteomic and non-proteomic that are indicative of breast cancer disease processes; and then (4) correlating (a) the presence or amount of said molecular markers and, with (b) clinicopathological data from said tissue sample other than the molecular markers of breast cancer disease processes. A kit of (1) a panel of antibodies; (2) one or more gene amplification assays; (3) first reagents to assist said antibodies with binding to tumor samples; (4) second reagents to assist in determining gene amplification; permits, when applied to a breast cancer patient's tumor tissue sample, (A) permits observation, and determination, of a numerical level of expression of each individual antibody, and gene amplification; whereupon (B) a computer algorithm, residing on a computer can calculate a prediction of treatment outcome for a specific treatment for breast cancer, or future risk of breast cancer progression.12-15-2011
20110145177HIERARCHICAL TEMPORAL MEMORY - Methods and systems for constructing biological-scale hierarchically structured cortical statistical memory systems utilizing fabrication technology and meta-stable switching devices. Learning content-addressable memory and statistical random access memory circuits are detailed. Additionally, local and global signal modulation of bottom-up and top-down processing for the initiation and direction of behavior is disclosed.06-16-2011
20110307423DISTRIBUTED DECISION TREE TRAINING - A computerized decision tree training system may include a distributed control processing unit configured to receive input of training data for training a decision tree. The system may further include a plurality of data batch processing units, each data batch processing unit being configured to evaluate each of a plurality of split functions of a decision tree for respective data batch of the training data, to thereby compute a partial histogram for each split function, for each datum in the data batch. The system may further include a plurality of node batch processing units configured to aggregate the associated partial histograms for each split function to form an aggregated histogram for each split function for each of a subset of frontier tree nodes and to determine a selected split function for each frontier tree node by computing the split function that produces highest information gain for the frontier tree node.12-15-2011
20110307426Personalized Health Risk Assessment For Critical Care - A method for providing a personalized health risk of a patient includes receiving training data corresponding to a plurality of patients and target data corresponding to a target patient; generating model data based on the training data according to an anomaly detection method; either determining whether the target data is anomalous with respect to the training data, or determining the extent to which the target data is anomalous with respect to the training data; and either indicating whether the target patient is at risk of the adverse outcome, or indicating the extent to which the target patient is at risk of the adverse outcome.12-15-2011
20110307425ORGANIZING SEARCH RESULTS - Many users make use of search engines to locate desired internet content by submitting search queries. For example, a user may search for photos, applications, websites, videos, documents, and/or information regarding people, places, and things. Unfortunately, search engines may provide a plethora of information that a user may be left to sift through to find relevant content. Accordingly, one or more systems and/or techniques for organizing search results are disclosed herein. In particular, user generated content, such as photos, may be retrieved based upon a search query. The user generated content may be grouped into clusters of user generated content having similar features. Search results of the search query may be obtained and organized based upon comparing the search results with the clusters. The organized search results and/or a table of content comprising the clusters may be presented to provide an enhanced user experience.12-15-2011
20110307424DETERMINATION OF TRAINING SET SIZE FOR A MACHINE LEARNING SYSTEM - Automated determination of a number of profiles for a training data set to be used in training a machine learning system for generating target function information from modeled profile parameters. In one embodiment, a first principal component analysis (PCA) is performed on a training data set, and a second PCA is performed on a combined data set which includes the training data set and a test data set. A test data set estimate is generated based on the first PCA transform and the second PCA matrix. The size of error between the test data set and the test data set estimate is used to determine whether a number of profiles associated with the training data set is sufficiently large for training a machine learning system to generate a library of spectral information.12-15-2011
20110307422EXPLORING DATA USING MULTIPLE MACHINE-LEARNING MODELS - A multiple model data exploration system and method for running multiple machine-learning models simultaneously to understand and explore data. Embodiments of the system and method allow a user to gain a greater understanding of the data and to gain new insights into their data. Embodiments of the system and method also allow a user to interactively explore the problem and to navigate different views of data. Many different classifier training and evaluation experiments are run simultaneously and results are obtained. The results are aggregated and visualized across each of the experiments to determine and understand how each example is classified for each different classifier. These results then are summarized in a variety of ways to allow users to obtain a greater understanding of the data both in terms of the individual examples themselves and features associated with the data.12-15-2011
20110307428Screening Information for a Coverage Model - It is disclosed to determine whether information useable for a generating/updating process that comprises generating and/or updating at least one model for a coverage area of a communication node shall be discarded or made available to said generating/updating process.12-15-2011
20090182690Detection and Classification of Light Sources Using a Diffraction Grating - A system mounted in a vehicle for classifying light sources. The system includes a lens and a spatial image sensor. The lens is adapted to provide an image of a light source on the spatial image sensor. A diffraction grating is disposed between the lens and the light source. The diffraction grating is adapted for providing a spectrum. A processor is configured for classifying the light source as belonging to a class selected from a plurality of classes of light sources expected to be found in the vicinity of the vehicle, wherein the spectrum is used for the classifying of the light source. Both the image and the spectrum may be used for classifying the light source or the spectrum is used for classifying the light source and the image is used for another driver assistance application.07-16-2009
20120041909METHOD TO CONFIGURE AN IMAGING DEVICE - A database contains variants of protocols for the operation of magnetic resonance tomographs as well as different types of magnetic resonance tomographs. Each variant contains parameter values and is associated with one of the types. In a training phase, relationships are determined between the parameters among one another and/or between the parameters and the associated types and are stored as patterns in a knowledge base. A protocol plan for the operation of a new magnetic resonance tomograph is created later in an application phase using the determined pattern. The method offers the advantage that the efficiency and quality of the automatic conversion of the protocols is improved. The improved quality of the protocol plan reduces operating time and costs for a manual post-processing of the protocols. Furthermore, a higher consistency of the protocols among one another is achieved both between product families and between individual configurations.02-16-2012
20120041911COMPUTER IMPLEMENTED SYSTEM AND METHOD FOR ASSESSING A NEUROPSYCHIATRIC CONDITION OF A HUMAN SUBJECT - A method for assessing a neuropsychiatric condition (such as, but not limited to, a risk that a subject may attempt to commit suicide or repeat an attempt to commit suicide, a risk that terminally ill patient is not being cared-for or treated according to the patient's true wishes, a risk that a subject may perform or repeat a criminal act and/or a harmful act, a risk of the subject having a psychiatric illness, and/or a risk of a subject feigning a psychiatric illness) may include a plurality of steps. A step may include receiving biomarker data associated from an analysis of the subject's biological sample and a step of receiving thought-marker data obtained pertaining to one or more of the subject's recorded thoughts, spoken words, transcribed speech, and writings. A step may include generating a biomarker score associated with the neuropsychiatric condition from the biomarker data. A step may include generating a thought-marker score associated with the neuropsychiatric condition from the thought-marker data. And a step may involve calculating a neuropsychiatric condition score based, at least in part, upon the biomarker score and the thought-marker score. Such method may be operating from one or more memory devices including computer-readable instructions configured to instruct a computerized system to perform the method, and the method may be operating on a computerized system including one or more computer servers (or other available devices) accessible over a computer network such as the Internet or over some other data network.02-16-2012
20120041910METHOD OF ESTABLISHING A PROCESS DECISION SUPPORT SYSTEM - A method of establishing a process decision support system. Decision support systems of the kind are used in manufacturing processes, particularly industrial manufacturing processes, to monitor the performance of the processes in view of controlling the processes in order to optimise process production and quality. The method includes collecting process data of a process, collecting operational data of a process, and fusing the process data and operational data to create a fused data set (such as a consolidated rule set) of the process upon which process decisions (such as control decisions) may be taken. The process data and operational data may be fused according to methods of rules-based knowledge fusion, mathematical knowledge fusion, or case-based reasoning knowledge fusion.02-16-2012
20120041908Predictive Radiosensitivity Network Model - This invention is a model that simulates the complexity of biological signaling in a cell in response to radiation therapy. Using gene expression profiles and radiation survival assays in an algorithm, a systems model was generated of the radiosensitivity network. The network consists of ten highly interconnected genetic hubs with significant signal redundancy. The model was validated with in vitro tests perturbing network components, correctly predicting radiation sensitivity 2/3 times. The model's clinical relevance was shown by linking clinical radiosensitivity targets to the model network. Clinical applications were confirmed by testing model predictions against clinical response to preoperative radiochemotherapy in patients with rectal or esophageal cancer.02-16-2012
20120041907Suggesting Connections to a User Based on an Expected Value of the Suggestion to the Social Networking System - To suggest new connections to a user of a social networking system, the system generates a set of candidate users to whom the user has not already formed a connection. The system determines the likelihood that the user will connect to each candidate user if suggested to do so, and it also computes the value to the social networking system if the user does connect to the candidate user. Then, the system computes an expected value score for each candidate user based on the corresponding likelihood and the value. The candidate users are ranked and the suggestions are provided to the user based on the candidate users' expected value scores. The social networking system can suggest other actions to a user in addition to forming a new connection with other users.02-16-2012
20120041906Supervised Nonnegative Matrix Factorization - Supervised kernel nonnegative matrix factorization generates a descriptive part-based representation of data, based on the concept of kernel nonnegative matrix factorization (kernel NMF) aided by the discriminative concept of graph embedding. An iterative procedure that optimizes suggested formulation based on Pareto optimization is presented. The present formulation removes any dependence on combined optimization schemes.02-16-2012
20120041905Supervised Nonnegative Matrix Factorization - Supervised nonnegative matrix factorization (SNMF) generates a descriptive part-based representation of data, based on the concept of nonnegative matrix factorization (NMF) aided by the discriminative concept of graph embedding. An iterative procedure that optimizes suggested formulation based on Pareto optimization is presented. The present formulation removes any dependence on combined optimization schemes. Analytical and empirical evidence is presented to show that SNMF has advantages over popular subspace learning techniques as well as current state-of-the-art techniques.02-16-2012
20120041904SYSTEM AND METHOD FOR MANAGING CONTINUED ATTENTION TO DISTANCE-LEARNING CONTENT - Management of a user's continued attention to distance learning content using a general purpose computer having a central processing unit and an operating system configured to run multiple program applications concurrently. A memory stores the distance learning content. A distance learning module comprises code executable on the central processing unit, as one of the multiple program applications. The distance learning module presents the distance learning content to a user and is operable to interrupt a presentation of the distance learning content in response to prescribed events concerning another one of the multiple program applications. A method executing on a computer that an concurrently run multiple applications identifies events concerning an application other than the distance learning application, processes the identified events so as to identify a prescribed event among the identified events, and interrupts the presentation of the distance learning content in response to the prescribed event.02-16-2012
20090171868Method and Apparatus for Early Termination in Training of Support Vector Machines - Disclosed is a method for early termination in training support vector machines. A support vector machine is iteratively trained based on training examples using an objective function having primal and dual formulations. At each iteration, a termination threshold is calculated based on the current SVM solution. The termination threshold increases with the number of training examples. The termination threshold can be calculated based on the observed variance of the loss for the current SVM solution. The termination threshold is compared to a duality gap between primal and dual formulations at the current SVM solution. When the duality gap is less than the termination threshold, the training is terminated.07-02-2009
20090171871COMBINATION MACHINE LEARNING ALGORITHMS FOR COMPUTER-AIDED DETECTION, REVIEW AND DIAGNOSIS - This invention utilizes a number of Computational Intelligence (CI) techniques with different learning methods in a computer-aided detection, review and diagnosis (CAD) device. Specifically, an unsupervised learning method is used for clustering of types of abnormal findings. Then a number of classifiers for each type of findings are trained with appropriate learning algorithms; and combined in three different manners to produce one classifier that can be operated at three different operating points. A fuzzy system is used for mapping the findings to diagnostic reports constructed using a formal language. Finally, the finding statistics is calculated based on Bayesian probability. During image review, the device provides the readers some insight as to how it derives its outputs. The output of the device can be updated in an interactive and progressive manner by a human reader (radiologist). The output from classification can be updated by the human, and is fed as input to the assessment task. Again the output from assessment can be updated by the human reader, and is fed as input for the machine to produce statistical information. If so configured, the interactive information can be added to an online database so that the device can adapt its future behavior based on the new information. 07-02-2009
20090171870System and method of feature selection for text classification using subspace sampling - An improved system and method is provided for feature selection for text classification using subspace sampling. A text classifier generator may be provided for selecting a small set of features using subspace sampling from the corpus of training data to train a text classifier for using the small set of features for classification of texts. To select the small set of features, a subspace of features from the corpus of training data may be randomly sampled according to a probability distribution over the set of features where a probability may be assigned to each of the features that is proportional to the square of the Euclidean norms of the rows of left singular vectors of a matrix of the features representing the corpus of training texts. The small set of features may classify texts using only the relevant features among a very large number of training features.07-02-2009
20090171869HOT TERM PREDICTION FOR CONTEXTUAL SHORTCUTS - Subject matter disclosed herein may relate to predicting hot terms, and may also relate to creating contextual shortcuts based, at least in part, on the predicted hot terms.07-02-2009
20090171867DETERMINING QUALITY OF TIER ASSIGNMENTS - Described herein is a method that includes receiving user history data and generating an indication of quality of a tier assignment used to store searchable digital items in a tiered storage system, wherein the indication is based at least in part upon a subset of the user history data. Also described herein is a system that includes a receiver component that receives user history data. The system further includes a quality indicator component that determines an indication of quality of a tier assignment used to store digital items that are retrievable by way of querying, wherein the quality indicator component generates the indication based at least in part upon a subset of the user history data and the tier assignment indicates where digital items are to be stored in a tiered storage system.07-02-2009
20120150775SYSTEM FOR SEMANTIC HOME NETWORK MANAGEMENT, CLOUD INFERENCE APPARATUS FOR SEMANTIC HOME NETWORK MANAGEMENT, SEMANTIC HOME NETWORK, AND SEMANTIC HOME NETWORK CONNECTION DEVICE - A semantic home network management system includes: a home network for collecting sensing information based on a state of a home network and analyzing the collected sensing information to configure semantic information; and a cloud inference apparatus for collecting and managing the semantic information provided from the home network through an internet protocol (IP) network, and applying an inference rule to the collected and managed semantic information to provide inference rule-applied semantic information to the home network. The inference rule is a rule that performs inference by recognizing at least one of a device state, a network state, a system state, and a service state.06-14-2012
20120150772Social Newsfeed Triage - A social newsfeed being delivered to a user is triaged. A personalized model is established which predicts the importance to the user of data elements within a current social newsfeed being delivered to the user. The personalized model is established based on implicit actions the user takes in response to receiving previous social newsfeeds. The personalized model is then used to triage the data elements within the current social newsfeed.06-14-2012
20090327171RECOGNIZING GESTURES FROM FOREARM EMG SIGNALS - A machine learning model is trained by instructing a user to perform proscribed gestures, sampling signals from EMG sensors arranged arbitrarily on the user's forearm with respect to locations of muscles in the forearm, extracting feature samples from the sampled signals, labeling the feature samples according to the corresponding gestures instructed to be performed, and training the machine learning model with the labeled feature samples. Subsequently, gestures may be recognized using the trained machine learning model by sampling signals from the EMG sensors, extracting from the signals unlabeled feature samples of a same type as those extracted during the training, passing the unlabeled feature samples to the machine learning model, and outputting from the machine learning model indicia of a gesture classified by the machine learning model.12-31-2009
20080201280MEDICAL ONTOLOGIES FOR MACHINE LEARNING AND DECISION SUPPORT - A medical ontology may be used for computer assisted clinical decision support. Multi-level and/or semantically grouped medical ontology is incorporated into a machine learning algorithm. The resulting machine-learnt algorithm outputs information to assist in clinical decisions. For example, a patient record is input to the algorithm. Based on the incorporated medical ontology, similarities are aggregated in different groups. An aggregate similarity of at least one group is a function of an aggregate similarity of at least another group. One or more similar patients and/or outcomes are identified based on similarity. Probability based outputs may be provided.08-21-2008
20090327170Methods of Clustering Gene and Protein Sequences - The invention relates to methods for clustering gene and protein sequences. In particular, it involves generation of networks of sequences where the interconnections are based upon a measure of similarity. The invention also provides methods of optimizing and improving the networks by re-wiring of the network based upon overlap of the nearest neighbors of given pairs of nodes. The invention further provides methods of identifying clusters of sequences within the networks and the optimized networks based upon the topology of the network. The clusters identified represent groups of sequences that are related by function and/or evolution. The invention has particular applicability in annotation of sequences in databases and identification of functional homologs which can be very useful for novel therapeutic and diagnostic targets based upon such targets belonging to a cluster or family that contains a known sequence such as a diagnostic sequence, antigen or other therapeutic target.12-31-2009
20120047098METHOD FOR COMPUTING AND STORING VORONOI DIAGRAMS, AND USES THEREFOR - A method of producing and storing a Voronoi diagram includes: a) selecting a desired site P02-23-2012
20120047097Secure Handling of Documents with Fields that Possibly Contain Restricted Information - A method, system and computer program product for processing documents containing restricted information. One aspect concerns identifying which sections of a document may be critical, non-critical or possibly critical.02-23-2012
20120047096METHOD AND APPARATUS FOR CLASSIFYING APPLICATIONS USING THE COLLECTIVE PROPERTIES OF NETWORK TRAFFIC - In one embodiment, the present disclosure is a method and apparatus for classifying applications using the collective properties of network traffic. In one embodiment, a method for classifying traffic in a communication network includes receiving a traffic activity graph, the traffic activity graph comprising a plurality of nodes interconnected by a plurality of edges, where each of the nodes represents an endpoint associated with the communication network and each of the edges represents traffic between a corresponding pair of the nodes, generating an initial set of inferences as to an application class associated with each of the edges, based on at least one measured statistic related to at least one traffic flow in the communication network, and refining the initial set of inferences based on a spatial distribution of the traffic flows, to produce a final traffic activity graph.02-23-2012
20120005133SYSTEM AND METHOD FOR MAPPING SS7 BEARER CHANNELS - A system and method for associating Signaling System 7 logical circuits and bearer channels are presented. The system may include an event detector configured to receive an SS7 signaling message on an SS7 signaling link, parse a logical circuit from the SS7 signaling message, receive an SS7 bearer channel, and detect a bearer channel event on the SS7 bearer channel. A statistical learning model block is configured to calculate a correlation confidence value between said bearer channel and said logical circuit. The method may include parsing a logical circuit ID from a signaling message on an SS7 signal link, identifying a bearer channel associated with a bearer event on a bearer circuit, and calculating a current correlation confidence value between the logical circuit ID and the bearer channel.01-05-2012
20120005132PREDICTING ESCALATION EVENTS DURING INFORMATION SEARCHING AND BROWSING - One or more techniques and/or systems are disclosed for predicting escalations in users' goals or concerns in web-based searching and browsing. One or more escalation features are extracted from a webpage. The one or more escalation features are run through a classifier trained to estimate a likelihood of escalation. An escalation likelihood result is generated from running the trained classifier using the extracted features. The escalation likelihood result can comprise an estimation that a subsequent search query will comprise an escalation when compared to a previous search query. The escalation likelihood result can also comprise an estimation that a subsequent webpage selection will comprise an escalation when compared to a previous webpage selection.01-05-2012
20110093417TOPICAL SENTIMENTS IN ELECTRONICALLY STORED COMMUNICATIONS - The present application presents methods for performing topical sentiment analysis on electronically stored communications employing fusion of polarity and topicality. The present application also provides methods for utilizing shallow NLP techniques to determine the polarity of an expression. The present application also provides a method for tuning a domain-specific polarity lexicon for use in the polarity determination. The present application also provides methods for computing a numeric metric of the aggregate opinion about some topic expressed in a set of expressions.04-21-2011
20110093416Systems, Methods, and Media for Performing Classification - Systems, methods, and media that: implement a boosted classifier having a plurality of weak hypotheses that produce a classification, each of the plurality of weak hypotheses having at least one weight; receive testing data; receive at least one piece of training data subsequently to receiving the testing data; calculate corrective terms for correcting a sum of weights of correctly classified training data and a sum of weights of incorrectly classified training data; calculate the sum of weights of correctly classified training data and the sum of weights of incorrectly classified training data based on the corrective terms; modify the at least one weight of at least one of the plurality of weak hypotheses in response to the at least one piece of training data based on the sum of weights to produce modified weights; and classify the testing data based on the modified weights to produce a classification.04-21-2011
20110093415CONTENT RECOMMENDATION APPARATUS AND METHOD - A content recommendation apparatus and method are provided. The content recommendation apparatus may record user history information of a user of a personal communication terminal where a web browsing service or mobile communication is possible. The user history information may be used to generate preference information of the user. Based on the preference information, content may be recommended to the user through a display based on a category type of the content such that different types of content are visually differentiated on the display.04-21-2011
20110093414SYSTEM AND METHOD FOR PHRASE IDENTIFICATION - A phrase identification system and method are provided. The method comprises: identifying one or more phrase candidates in the electronic document; selecting one of the phrase candidates; numerically representing features of the selected phrase candidates to obtain a numeric feature representation associated with that phrase candidate; and inputting the numeric feature representation into a machine learning classifier, the machine learning classifier being configured to determine, based on each numeric feature representation, whether the phrase candidate associated with that numeric feature representation is a phrase.04-21-2011
20120011083Product-Centric Automatic Software Identification in z/OS Systems - A software identification manager selects a unique software module that corresponds to a single knowledge base software product. Next, the software identification manager determines that one or more software modules included in a raw inventory corresponds to the unique software module and, in turn, the software identification manager includes the single knowledge base software product into a refined group of knowledge base software products. The software identification manager then matches one of the knowledge base software products included in the refined group of knowledge base software products to one of the related raw inventory software modules. Once the software identification manager identifies a match, the software identification manager stores a module identification entry in a storage area and associates the matched raw inventory software module to the matched knowledge base software product.01-12-2012
20120011082GOVERNANCE OF MODELING SYSTEMS - A governing modeling system maintains information associated with an analytic model. One or more policies may be defined that are associated with the analytic model and one or more instances of the analytic model. The system may monitor the analytic model and the one or more instances of the analytic model based on at least some of the information, the one or more policies associated with the analytic model and the one or more policies associated with the one or more instances of the analytic model.01-12-2012
20110167027INFORMATION ANALYSIS APPARATUS, INFORMATION ANALYSIS METHOD, AND COMPUTER-READABLE RECORDING MEDIUM - An information analysis apparatus, an information analysis method, and a program are provided that enable target information to be determined in units of single sentences, rather than in units of plural sentences, while taking into consideration the tendency of appearance of the target information. An information analysis apparatus 07-07-2011
20110167026SYSTEMS AND METHODS FOR PROVIDING EXTENSIBLE ELECTRONIC LEARNING SYSTEMS - An extensible electronic learning system having at least one learning management system having a learning management processor and a learning management memory operatively coupled thereto, said processor programmed for executing at least one learning management service and providing at least one extensible integration module. Each extensible integration module includes a predefined vendor services interface comprising at least one vendor services definition, and a vendor configuration upload component for receiving vendor configuration settings about at least one vendor. The at least one vendor having a vendor processor and a vendor memory operatively coupled thereto, said vendor processor programmed for executing a least one vendor services, the at least one of said vendor services implementing the at least one of said vendor service definition, and providing at least one vendor integration module, each vendor integration module comprising the predefined vendor services interface and the vendor configuration settings. The vendor configuration settings are received by the extensible integration module such that the learning management system may request the at least one of said vendor services based on the extensible integration module.07-07-2011
20110167025SYSTEMS AND METHODS FOR PARAMETER ADAPTATION - A method of parameter adaptation is used to modify the parameters of a model to improve model performance. The model separately estimates the contribution of each model parameter to the prediction error. It achieves this by transforming to the time-scale plane the vectors of output sensitivities with respect to model parameters and then identifying the regions within the time-scale plane at which the dynamic effect of individual model parameters is dominant on the output. The method then attributes the prediction error in these regions to the deviation of a single parameter from its true value as the basis of estimating individual parametric errors. The proposed Signature Isolation Method (PARSIM) then uses these estimates to adapt individual model parameters independently of the others, implementing, in effect, concurrent adaptation of individual parameters by the Newton-Raphson method in the time-scale plane. The Signature Isolation Method has been found to have an adaptation precision comparable to that of the Gauss-Newton method for noise-free cases. However, it surpasses the Gauss-Newton method in precision when the output is contaminated with noise or when the measurements are generated by inadequate excitation.07-07-2011
20120066161SIMPLIFIED ALGORITHM FOR ABNORMAL SITUATION PREVENTION IN LOAD FOLLOWING APPLICATIONS INCLUDING PLUGGED LINE DIAGNOSTICS IN A DYNAMIC PROCESS - Systems and methods are provided for detecting abnormal conditions and preventing abnormal situations from occurring in controlled processes. Statistical signatures of a monitored variable are modeled as a function of the statistical signatures of a load variable. The statistical signatures of the monitored variable may be modeled according to an extensible regression model or a simplified load following algorithm. The systems and methods may be advantageously applied to detect plugged impulse lines in a differential pressure flow measuring device.03-15-2012
20120066160PROBABILISTIC TREE-STRUCTURED LEARNING SYSTEM FOR EXTRACTING CONTACT DATA FROM QUOTES - Systems and methods for updating data stored in a database, such as contact information. An input string is obtained through a search for timely material associated with the stored contact. The input string is parsed using probabilistic tendencies to extract entities corresponding to those stored with the contact. Secondary entities are used to assist in the identification of the primary entities. The contact is then updated (or added if new) using the extracted primary entities.03-15-2012
20130013534HARDWARE-ASSISTED APPROACH FOR LOCAL TRIANGLE COUNTING IN GRAPHS - A method and apparatus are provided for hardware-assisted local triangle counting in a graph. The method includes converting vertex relationships of the graph into rule patterns. The method also includes compiling the rule patterns into a binary file, wherein the rule patterns are organized into a finite state machine. The method further includes loading at least a part of the binary file and a search string to be compared there against into a hardware pattern matching accelerator. The method additionally includes receiving a number of matching outputs from the pattern matching accelerator.01-10-2013
20120209797System And Method For Facilitating Evergreen Discovery Of Digital Information - A computer-implemented system and method for facilitating evergreen discovery of digital information is provided. A hierarchy of topics for topically-limited subject areas is defined. Seed words characteristic of each topic are selected. Training material from the digital information that corresponds to the respective subject area of each of the topics is designated. Candidate topic models are formed from the seed words. Each candidate topic model includes a pattern evaluable against the digital information. An ability of each of the candidate topic models to identify such digital information matching the candidate topic model's topic is tested by matching the pattern in the candidate topic model to the training material. The candidate topic model for each topic that includes the highest abilities with respect to the topic in performance, simplicity and bias is chosen. An evergreen index is formed by pairing the chosen candidate topic model to each topic in the hierarchy.08-16-2012
20120209796ATTENTION FOCUSING MODEL FOR NEXTING BASED ON LEARNING AND REASONING - A system and method for nexting is presented. The method comprises computing an expected event, observing a new event, when the expected event matches the new event, processing the new event and performing action in accordance with given concepts, when the expected event does not match the new event and the new event can be explained based on the given concepts, processing the new event and performing action in accordance with the given concepts, and when the expected event does not match the new event and the new event cannot be explained based on the given concepts, employing learning mechanism and performing action decided on by the learning mechanism. In one aspect, the method comprises generating new concepts using reasoning or learning. In one aspect, the method comprises converting sensed numerical data into events of interest via the application of learned functions operating on the numerical data.08-16-2012
20120016822INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM - There is provided an information processing method including inputting a feature quantity vector and an objective variable corresponding to the feature quantity vector, generating a basis function for outputting a scalar quantity by mapping the feature quantity vector, mapping the feature quantity vector using the basis function and calculating the scalar quantity corresponding to the feature quantity vector, evaluating whether or not the basis function used to calculate the scalar quantity is useful for estimating the objective variable using the objective variable along with the scalar quantity and the feature quantity vector corresponding to the scalar quantity, generating an estimation function for estimating the objective variable from the scalar quantity by machine learning on the basis of the scalar quantity and the objective variable corresponding to the scalar quantity using the basis function evaluated to be useful, and outputting the estimation function.01-19-2012
20120016820Stochastic search strategies for multimedia resource discovery and retrieval system - A method is described for applying distributed stochastic optimization techniques of evolutionary computation using a plurality of servers and a plurality of clients machines being connected via a computer network such as the Internet. The stochastic optimization techniques of evolutionary computation seek to optimize a populations of individuals against one or more predetermined fitness criteria when applied to solving solve the network routing problem coupled with one or more information retrieval problems. The field of evolutionary computation encompasses stochastic optimization techniques, such as randomized search strategies, in the form of evolutionary strategies (ES), evolutionary programming (EP), genetic algorithms (GA), classifier systems, evolvable hardware (EHW), and genetic programming (GP). The stochastic optimization component objectives of the multimedia resource discovery and retrieval systems includes maximization of resource utilization and of overall LAN throughput.01-19-2012
20120016824METHOD FOR COMPUTER-ASSISTED ANALYZING OF A TECHNICAL SYSTEM - A method for computer-assisted analyzing of a technical system is provided. The technical system is described by a case base including multiple cases, each case including a state vector with a number of attributes, the state vector referring to an operation state of the technical system, wherein a class from a number of classes is assigned to each case, each class referring to an operation condition of the technical system. Each case is processed by extracting a local information vector depending on the classes of one or more neighboring cases in the case base, the neighboring cases being similar to the case being processed according to a neighborhood measure. Subsequently, machine learning of a classification is performed based on the extracted local information vectors of the cases in the case base, resulting in a learned adaptation function providing a class depending on a local information vector extracted for a case.01-19-2012
20120016823DATA COMPRESSION METHOD FOR A CLASSIFIER - A method of classifying a sample of values related to the use of a server, comprises the steps of recording, by the server, use events in a log; configuring a classifier tool with a behavioral model formed of a weighted list of parameters; establishing the sample of values from the log and supplying it as parameters to a classifier tool, whereby the classifier tool calculates a score representative of the adequacy of the sample to a target category; reading recent use events saved in the log and aggregating them over basic time intervals; storing in a database the aggregation result obtained for each basic interval in a distinct record of a first group; aggregating, when the number of records of the first group reaches a threshold, the contents of the records of the first group in a distinct record of a second group of the database; and establishing the sample of values supplied to the classifier tool from the contents of records of the database.01-19-2012
20120016818Classification of Biological Samples Using Spectroscopic Analysis - A method and system is described for rapidly classifying a sample of a biological fluid, comprising obtaining a spectrum of the biological fluid in response to excitation of the sample in a specified frequency range, and applying a multivariate classifier to one or more spectral regions of the spectrum to classify the biological sample into one class in a set of classes, the classes comprising at least two disease states having similar clinical symptoms. Methods and systems for developing the classifiers are also described. In one example the classification uses a vibrational spectrometer (01-19-2012
20120016817Predicting Life Changes of Members of a Social Networking System - To predict a life change event for a user of the social networking system, such as a change in marital status, relationship status, employment status, etc., the disclosed system generates a training set of data comprising historical data of other users who have gone through a life change event. The system uses the training set data to generate a prediction algorithm using machine learning models. Furthermore, the system inputs the user data to the prediction algorithm to retrieve a prediction of whether the user will undergo one or more life change events. The system updates the user's profile to indicate the life change event and provides advertisements to the user responsive to the prediction of one or more life change events.01-19-2012
20120016819Distributed multimedia document indexing strategies - A method for a system that indexes/ranks/clusters multimedia documents using hybrids of information retrieval algorithms and the stochastic optimization techniques of evolutionary computation (EC) that optimizes parameter sets comprising of object parameters. The method creates a plurality of individual parameter sets, the parameter sets comprising information sharing system object parameters for describing a model, structure, shape, design, process, search query set, or dynamic search space to be optimized and setting the initial population as a current (static parent) population. These parameters are required to filter, organize, and index any large-scale data set—information stored on a single computer, a local area network (LAN), and a wide area network (WAN) that encompasses the whole Internet—that may consists of constantly fluctuating information content over relatively short periods of time01-19-2012
20120254079Serendipitous Recommendations System and Method - A computer-implemented serendipitous recommendations system and method generates recommendations for delivery to system users in accordance with settings of desired levels of serendipity, including serendipity levels established through use of serendipity tuning controls operable by users. The recommendations are informed by an interest affinity anomaly function that identifies contrasting interest affinities between recommendation recipients and other users. Explanations may be generated that provide reasons as to why a recommendation was delivered to a user, and the explanation may include a selection of phrases that are influenced by a serendipity level setting, and may include an expression of a level of confidence with regard to the recommendation.10-04-2012
20120254078MARKOV MODELING OF SERVICE USAGE PATTERNS - A system for analyzing service usage utilizing Markov models. Records of client requests to the service are extracted from at least one log. The records are grouped by client and sorted by timestamp. A pattern of requests that form an action is detected. Each action has a time. A probability is calculated of a transition from a precedent action to a subsequent action, where the precedent action has a time prior to the subsequent action. A delay time is also calculated between a precedent action and a subsequent action. A probability is calculated for a delay time, such as the likelihood that a delay from a precedent action to a subsequent action will fall within a given time interval.10-04-2012
20120023045Recommender System with Training Function Based on Non-Random Missing Data - A processing device of an information processing system is operative to obtain observed feedback data, to construct a model that accounts for both the observed feedback data and additional feedback data that is missing from the observed feedback data, to optimize one or more parameters of the model using a training objective function, and to generate a list of recommended items for a given user based on the optimized model. In illustrative embodiments, the missing feedback data comprises data that is missing not at random (MNAR), and the model comprises a matrix factorization model. The processing device may implement a recommender system comprising a training module coupled to a recommendation module.01-26-2012
20120023042CONFIDENCE LEVEL GENERATOR FOR BAYESIAN NETWORK - A system includes a computer implemented Bayesian diagnostic system. The diagnostic system includes an inferencing engine and a conditional probability table that forms the basis for Bayesian inferences once the diagnostic system is trained. Each inference includes a diagnosis and associated probability of the diagnosis. A confidence generator receives the inferences, and generates a confidence measure for each inference.01-26-2012
20120023044Issue Resolution in Expert Networks - Techniques are provided for improved issue resolution in an expert network. For example, a method comprises the following steps. Information is extracted comprising: content of one or more historical records associated with resolutions of one or more previous issues; and transfer routing sequences indicating routes through routing entities in an expert network that the one or more previous issues passed in order to be respectively resolved;. A model is computed based on at least a portion of the extracted information, wherein the computed model statistically captures one or more ticket transfer patterns among routing entities in the expert network. One or more future issue resolution routing recommendations are determined based on at least one of the one or more ticket transfer patterns captured by the computed model.01-26-2012
20120023041SYSTEM AND METHOD FOR PREDICTIVE NETWORK MONITORING - A system and a method for at least predicting a trend toward a reduction in performance of a computer and/or a computer network. Preferably, the system and method is able to predict a trend toward a potential failure of a computer and/or a computer network.01-26-2012
20120059779Personalized Health Risk Assessment For Critical Care - A method for assessing whether a patient is at risk of developing a clinical condition includes receiving training data representing a set of patient-related variables for each of a plurality of patients; generating model data based on the received training data; receiving target data representing the set of patient-related variables for a target patient; determining a risk level for the target patient of developing the clinical condition; and indicating the risk level of the target patient, where the set of patient-related variables consists of a first set of variables when the clinical condition is a mortality condition and a second set of variables when the clinical condition is a morbidity condition.03-08-2012
20120059778SELF-IMPROVING CLASSIFICATION SYSTEM - A self-improving classification system classifies specimens based on class identifiers. The system stores specimen profiles in a database that is updated with additional specimen profiles and with follow-up data that corrects classification of specimens that were initially incorrectly classified. Algorithms use the updated database to discover new class identifiers, modify thresholds of known class identifiers, and drop unnecessary class identifiers to improve classification of specimens.03-08-2012
20120059777CHARACTERIZING DATASETS USING SAMPLING, WEIGHTING, AND APPROXIMATION OF AN EIGENDECOMPOSITION - A method, a system, and a computer-readable medium are provided for characterizing a dataset. A representative dataset is defined from a dataset by a computing device. The representative dataset includes a first plurality of data points and the dataset includes a second plurality of data points. The number of the first plurality of data points is less than the number of the second plurality of data points. The data point is added to the representative dataset if a minimum distance between the data point and each data point of the representative dataset is greater than a sampling parameter. The data point is added to a refinement dataset if the minimum distance between the data point and each data point of the representative dataset is less than the sampling parameter and greater than half the sampling parameter. A weighting matrix is defined by the computing device that includes a weight value calculated for each of the first plurality of data points based on a determined number of the second plurality of data points associated with a respective data point of the first plurality of data points. The weight value for a closest data point of the representative dataset is updated if the minimum distance between the data point and each data point of the representative dataset is less than half the sampling parameter. A machine learning algorithm is executed by the computing device using the defined representative dataset and the defined weighting matrix applied in an approximation for a computation of a full kernel matrix of the dataset to generate a parameter characterizing the dataset.03-08-2012
20120158621STRUCTURED CROSS-LINGUAL RELEVANCE FEEDBACK FOR ENHANCING SEARCH RESULTS - A “Cross-Lingual Unified Relevance Model” provides a feedback model that improves a machine-learned ranker for a language with few training resources, using feedback from a more complete ranker for a language that has more training resources. The model focuses on linguistically non-local queries, such as “world cup” (English language/U.S. market) and “copa mundial” (Spanish language/Mexican market), that have similar user intent in different languages and markets or regions, thus allowing the low-resource ranker to receive direct relevance feedback from the high-resource ranker. Among other things, the Cross-Lingual Unified Relevance Model differs from conventional relevancy-based techniques by incorporating both query- and document-level features. More specifically, the Cross-Lingual Unified Relevance Model generalizes existing cross-lingual feedback models, incorporating both query expansion and document re-ranking to further amplify the signal from the high-resource ranker to enable a learning to rank approach based on appropriately labeled training data.06-21-2012
20120158623VISUALIZING MACHINE LEARNING ACCURACY - The claimed subject matter provides a method for visualizing machine learning accuracy. The method includes receiving a plurality of training instances for the machine learning system. The method also includes receiving a plurality of results for the machine learning system. The plurality of results corresponds to the plurality of training instances. The method further includes providing an interactive representation of the training instances and the results. The interactive representation supports identifying inaccuracies of the machine learning system attributable to the training instances, the features used to obtain a featurized form of the training instance, and/or a model implemented by the machine learning system.06-21-2012
20120158624PREDICTIVE MODELING - A predictive analysis generates a predictive model (Padj(Y|X)) based on two separate pieces of information, 06-21-2012
20120158622INTERACTIVE RECOMMENDATIONS - An interactive recommendation system generates one or more recommendations (e.g., recommended products, travel destinations, etc.) for a user based on a recommendation model. The recommendation model includes one or more criteria that are used to analyze a datastore of user characteristics (e.g., a user's age, location, past online behavior, etc.) and generate one or more recommendations based thereon. The interactive recommendation system further presents a user interface that allows the user to interactively modify the criteria of the recommendation model and to apply the modified recommendation model to the datastore in order to generate one or more modified recommendations. In this manner, for example, the user can customize the recommendations he or she receives by interacting with the recommendation system to modify the recommendation model used to generate such recommendations.06-21-2012
20120158619OPTIMAL RULE SET MANAGEMENT - Systems, methods, and computer products for optimally managing large rule sets are disclosed. Rule dependencies of rules within a set of rules may be determined as a function of rules execution frequency data generated from applying the rules over a data set. The rules within the set of rules may be clustered into rules clusters based on the determined rule dependencies, in which the rules clusters comprise disjoint subsets of the rules within the set of rules. Cluster frequency data for the rules clusters may be used to arrive at an optimal ordering.06-21-2012
20120158625Creating and Processing a Data Rule - A data rule is created and processed by receiving an expression defining a logic of a rule and at least one logical variable, creating a rule definition including the expression and the at least one logical variable for binding each logical variable of the rule with at least one column, associating a characteristic enabling comparison of columns with a first logical variable of the rule definition, and storing the characteristic as part of the rule definition.06-21-2012
20120158620HUMAN-ASSISTED TRAINING OF AUTOMATED CLASSIFIERS - Many computing scenarios involve the classification of content items within one or more categories. The content item set may be too large for humans to classify, but an automated classifier (e.g., an artificial neural network) may not be able to classify all content items with acceptable accuracy. Instead, the automated classifier may calculate a classification confidence while classifying respective content items. Content items having a low classification confidence may be sent to a human classifier, and may be added, along with the categories identified by the human classifier, to a training set. The automated classifier may then be retrained using the training set, thereby incrementally improving the classification confidence of the automated classifier while conserving the involvement of human classifiers. Additionally, human classifiers may be rewarded for classifying the content items, and the costs of such rewards may be considered while selecting content items for the training set.06-21-2012
20120072381Method and Apparatus for Segmenting Context Information - An approach is provided for segmenting context information. A context segmenting platform determines context information associated with a device. The context segmenting platform determines context information associated with a device. The context segmenting platform then determines one or more context patterns based, at least in part, on the context information and determines one or more transition points between the one or more context patterns. Based, at least in part, on the one or more transition points, the context segmenting platform determines to segment the context information.03-22-2012
20120072380REGULAR EXPRESSION MATCHING USING TCAMS FOR NETWORK INTRUSION DETECTION - A method is provided for implementing regular expression matching using ternary content-addressable memory devices. The method includes: receiving a set of regular expressions (REs) that specify data elements to be extracted from data packets; constructing a deterministic finite automaton (DFA) from the set of regular expressions; building a state transition table for each node of the deterministic finite automaton; combining the state transition tables into a single lookup table; and instantiating the lookup table in a ternary content-addressable memory device. Additional techniques are provided to reduce the TCAM space and improve RE matching speed.03-22-2012
20110087626PRODUCT CLASSIFICATION IN PROCUREMENT SYSTEMS - Various embodiments provide solutions to assist in the classification of products in a procurement system. The tools provided by various embodiments include, without limitation, methods, systems, and/or software products. Merely by way of example, a method might comprise one or more procedures, any or all of which are executed by a computer system. Correspondingly, an embodiment might provide a computer system configured with instructions to perform one or more procedures in accordance with methods provided by various other embodiments. Similarly, a computer program might comprise a set of instructions that are executable by a computer system (and/or a processor therein) to perform such operations. In many cases, such software programs are encoded on physical and/or tangible computer readable media (such as, to name but a few examples, optical media, magnetic media, and/or the like).04-14-2011
20120078822METHOD AND APPARATUS FOR PROVIDING A FRAMEWORK FOR GENERATING RECOMMEDATION MODELS - An approach is provided for providing a framework for generating recommendation models. A recommendation engine receives a request, at the recommendation engine, for generating a recommendation model for an application, wherein the recommendation engine is applicable to a plurality of applications. Next, the recommendation engine determines to retrieve rating information from on one or more profiles associated with the application, one or more other applications, or a combination thereof. Then, the recommendation engine determines to generate the recommendation model based, at least in part, on the rating information.03-29-2012
20120158618REMOTE NON-INTRUSIVE OCCUPANT SPACE MONITORING SYSTEM - A system for remote non-intrusive occupant space monitoring. The system may have sensors and other mechanisms for non-intrusively obtaining information by capturing utility and communication signals, images, light, sound, environmental factors, background information, and so on, about a space and its occupants. The obtained information may be locally or remotely analyzed and modeled by a processor. Models of buildings, behavior, and power systems from the processor may be compared with pre-defined models to infer further information about the space and its occupants. Also, behavioral information may be obtained, inferred and/or learned. The models may be updated with the obtained, inferred and learned information.06-21-2012
20110066578METHOD AND SYSTEM FOR PARALLEL STATISTICAL INFERENCE ON HIGHLY PARALLEL PLATFORMS - Methods for faster statistical inference in computation based recognition problems on highly parallel processors with multiple cores on-a-chip are disclosed, which include: selectively flattening levels of the recognition network to improve inference speed (improving the recognition model); selectively duplicating parts of the recognition network to minimize a critical section in atomic accesses to as few as one atomic instruction (improving the recognition procedure); and combining weight and source port into one 32-bit word to minimize the number of atomic operations. These methods have been implemented on an NVIDIA GTX 280 processor in a Large Vocabulary Continuous Speech Recognition (LVCSR) embodiment, and achieve more than a 10× speed up compared to a highly optimized sequential implementation on an Intel Core i7 processor.03-17-2011
20120084237DATA PROCESSING DEVICE, DATA PROCESSING METHOD, AND PROGRAM - A data processing device includes a state value calculation unit which calculates a state value of which the value increases as much as a state with a high transition probability for each state of the state transition model, an action value calculation unit which calculates an action value, of which the value increases as a transition probability increases for each state of the state transition model and each action that the agent can perform, a target state setting unit which sets a state with great unevenness in the action value among states of the state transition model to a target state that is the target to reach by action performed by the agent, and an action selection unit which selects an action of the agent so as to move toward the target state.04-05-2012
20110106736SYSTEM AND METHOD FOR INTUITIVE USER INTERACTION - The disclosed method and apparatus provide prediction and suggestion of proposed actions a user of an electronic device is likely to want to do, at certain circumstances. The actions take into account historical activities made by the user, as well as incoming events, environmental data, external data, or any other source of information. Proposing the actions may be done by one or more engines, each relating to one or more aspects of the device, actions, events, activities, preferences and the like. The actions proposed by all engines are merged and prioritized, and presented to a user. The options are presented to a user in a manner that enables activation of any of the options, with the relevant settings and parameters.05-05-2011
20110106733SYSTEM AND METHOD FOR PATIENT-MANAGED ORAL ANTICOAGULANT THERAPY - A system for facilitating patient-managed anticoagulant therapy having patient portals for receiving patient information associated with a current case. Each patient portal has a patient portal processor, patient portal memory, and an output device, and the patient information including current factors affecting patient reaction to anticoagulant therapy. The system also has a case repository connected to each patient portal, the case repository having a case repository processor and a case repository memory storing previous cases, each case having previous factors affecting patient reaction to anticoagulant therapy and at least one solution. The patient portal processor and/or the case repository processor are programmed for selecting at least one relevant case similar to the current case and provide the solution to that case for application to the current case. A similarity metric is applied to each previous case to determine the similarity of each previous case to the current case.05-05-2011
20110106732METHOD FOR CATEGORIZING LINKED DOCUMENTS BY CO-TRAINED LABEL EXPANSION - Systems and methods are described that facilitate categorizing a group of linked web pages. A plurality of web pages each contains at least one link to another page within the group. A feature analyzer evaluates features associated with the one or more web pages to identify content, layout, links and/or metadata associated with the one or more web pages and identifies features that are labeled and features that are unlabeled. A graphing component creates a vector associated with each web page feature wherein vectors for unlabeled features are determined by their graphical proximity to features that are labeled. A co-training component receives the graph of vectors from the graphing component and leverages the disparate web page features to categorize each aspect of each feature of the page. A page categorizer receives aspect categorization information from the co-training component and categorizes the web page based at least upon this information.05-05-2011
20110106735RECURSIVE FEATURE ELIMINATION METHOD USING SUPPORT VECTOR MACHINES - Identification of a determinative subset of features from within a group of features is performed by training a support vector machine using training samples with class labels to determine a value of each feature, where features are removed based on their the value. One or more features having the smallest values are removed and an updated kernel matrix is generated using the remaining features. The process is repeated until a predetermined number of features remain which are capable of accurately separating the data into different classes. In some embodiments, features are eliminated by a ranking criterion based on a Lagrange multiplier corresponding to each training sample.05-05-2011
20120317059SYSTEM AND METHOD FOR SPACE AND RESOURCE OPTIMIZATION - Methods and systems for space and resource optimization are disclosed, including a method comprising receiving a plurality of inputs, transforming the plurality of inputs into at least one or more of (but not limited to) an algorithmic graph and a structural graph based on a domain-specific area using a computer processor. The method further includes creating and applying heuristics for parallelization, performing an optimization run, and analyzing an optimal result produced by the optimization run.12-13-2012
20100094786Smoothed Sarsa: Reinforcement Learning for Robot Delivery Tasks - The present invention provides a method for learning a policy used by a computing system to perform a task, such delivery of one or more objects by the computing system. During a first time interval, the computing system determines a first state, a first action and a first reward value. As the computing system determines different states, actions and reward values during subsequent time intervals, a state description identifying the current sate, the current action, the current reward and a predicted action is stored. Responsive to a variance of a stored state description falling below a threshold value, the stored state description is used to modify one or more weights in the policy associated with the first state.04-15-2010
20100094785SURVIVAL ANALYSIS SYSTEM, SURVIVAL ANALYSIS METHOD, AND SURVIVAL ANALYSIS PROGRAM - Disclosed is a survival analysis system for determining an estimated time until an event occurs on the basis of a group of cases each including at least one attribute value indicating a feature value of a case and information on the measured actual time until an event occurs. The survival analysis system includes: an estimator creating section for creating an estimator for estimating whether or not an event occurs according to the attributes of the group of cases for each actual time; an estimator selecting section for judging whether or not the estimator meets a predetermined selection condition and to selecting an estimator used for calculating the estimated time; and a time calculating section for calculating the estimated time by using the estimator selected by the estimator selecting section.04-15-2010
20100094784GENERALIZED KERNEL LEARNING IN SUPPORT VECTOR REGRESSION - A generalized kernel learning system and method for learning a wide variety of kernels for use in a support vector regression (SVR) technique. Embodiments of the generalized kernel learning system and method learn nearly any possible kernel, subject to minor constraints. The learned kernel then is used to obtain a desired function, which is a function that closely fits training data and has a desired simplicity. Embodiments of the generalized kernel learning method include inputting the training data, reformulating a and a standard SVM ε-SVR primal formulation for a single kernel as two reformulated primal cost functions for multiple kernels, and then reformulating one of the two reformulated primal cost functions as a reformulated dual cost function. A plurality of different regularizer and kernel combinations is evaluated using the reformulated dual cost function, and it is determined which regularizer and kernel combination yields the desired function.04-15-2010
20100094782Information Processing Apparatus, Information Processing Method, and Program - The present invention relates to an information processing apparatus, an information processing method, and a program capable of quickly and accurately creating an algorithm for extracting features from content data such as song data. A feature extraction algorithm creation apparatus includes: a low-level feature extraction expression list creation section 21 that creates as many as “n” low-level feature extraction expression lists each constituted by “m” low-level feature extraction expressions; a low-level feature computation section 24 that substitutes input data of “j” songs into “n” low-level feature extraction expression lists so as to acquire “n” combinations of “m” low-level features corresponding to each input data item; and a high-level feature extraction expression learning section 25 that estimates high-level feature extraction expressions through learning based on training data corresponding to “n” low-level feature outputs (“k” high-level features corresponding to each of “j” songs) . This invention can be applied to systems for acquiring high-level features of songs and videos.04-15-2010
20120084236Recording medium storing decision tree generating program, decision tree generation method and decision tree generating apparatus - A constraint condition DB storing a constraint condition that stipulates a structure of a decision tree is referenced, and a decision tree is generated from a case set where values of a plurality of attributes and a conclusion are associated with one another so that the structure of the decision tree, which is stipulated by the constraint condition, is satisfied. Accordingly, for example, even if a new case is added to the case set, a basic structure of the decision tree is succeeded by the constraint condition, thereby avoiding a situation where an operator needs to significantly modify the decision tree. Therefore, operations of modifying the decision tree by the operator can be reduced.04-05-2012
20090132445GENERALIZED REDUCED ERROR LOGISTIC REGRESSION METHOD - A machine classification learning method titled Generalized Reduced Error Logistic Regression (RELR) is presented. The method overcomes significant limitations in prior art logistic regression and other machine classification learning methods. The method is applicable to all current applications of logistic regression, but has significantly greater accuracy using smaller sample sizes and larger numbers of input variables than other machine classification learning methods including prior art logistic regression.05-21-2009
20090132447Support Vector Machines Processing System - An implementation of SVM functionality improves efficiency, time consumption, and data security, reduces the parameter tuning challenges presented to the inexperienced user, and reduces the computational costs of building SVM models. A system for support vector machine processing comprises data stored in the system, a client application programming interface operable to provide an interface to client software, a build unit operable to build a support vector machine model on at least a portion of the data stored in the system, the portion of the data selected using a stratified sampling method with respect to a target distribution, an apply unit operable to apply the support vector machine model using the data stored in the system.05-21-2009
20090132443Methods and Devices for Analyzing Lipoproteins - The disclosure describes methods, systems, and devices for analysis of lipoproteins and for diagnosing and/or determining risk of cardiovascular disease. In some embodiments, lipoproteins are separated by electrophoretically using a micro-channel device, and the data are analyzed using an adaptive method such as a neural network.05-21-2009
20110119212EXPERT SYSTEM FOR DETERMINING PATIENT TREATMENT RESPONSE - A medical digital expert system to predict a patient's response to a variety of treatments (using pre-treatment information) is described. The system utilizes data fusion, advanced signal/information processing and machine learning/inference methodologies and technologies to integrate and explore diverse sets of attributes, parameters and information that are available to select the optimal treatment choice for an individual or for a subset of individuals suffering from any illness or disease including psychiatric, mental or neurological disorders and illnesses. The methodology and system can also be used to determine or confirm medical diagnosis, estimate the level, index, severity or critical medical parameters of the illness or condition, or provide a list of likely diagnoses for an individual suffering/experiencing any illness, disorder or condition.05-19-2011
20120221499WORKLOAD LEARNING IN DATA REPLICATION ENVIRONMENTS - A method for replicating I/O performance in data replication environments, such as PPRC environments, is described. In selected embodiments, such a method includes monitoring I/O workload at a primary storage device over a period of time, such as a period of hours, days, or months. The method then generates learning data at the primary storage device describing the I/O workload over the selected time period. The learning data is replicated from the primary storage device to a secondary storage device. The method uses the learning data to optimize the secondary storage device to handle the I/O workload of the primary storage device. This will enable the secondary storage device to provide substantially the same I/O performance as the primary storage device in the event a failover occurs.08-30-2012
20120221498AGGREGATING AND NORMALIZING ENTERTAINMENT MEDIA - Disclosed are methods for making disparate entertainment media content (e.g., television or movies) from multiple sources available through a single interface of a user device. Content of varying data formats from multiple data sources are aggregated. Classifications of the media data are created which can include assigning content into clusters. The data are normalized, and attributes of the data are curated. Features also are provided to automatically synchronize, obtain, and update media content on the media sources and on client devices. Various ways of handling data aggregation and normalization issues associated with compiling media data also are described.08-30-2012
20120221496Text Classification With Confidence Grading - A computer implemented method and system is provided for classifying a document. A classifier is trained using training documents. A list of first words is obtained from the training documents. A prior probability is determined for each class of multiple classes. Conditional probabilities are calculated for the first words for each class. Confidence thresholds are determined. Confidence grades are defined for the classes using the confidence thresholds. A list of second words is obtained from the document. Conditional probabilities for the list of second words are determined from the calculated conditional probabilities for the list of first words. A posterior probability is calculated for each of the classes and compared with the determined confidence thresholds. Each class is assigned to one of the defined confidence grades based on the comparison. The document is assigned to one of the classes based on the posterior probability and the assigned confidence grades.08-30-2012
20120221495DIGITAL WEIGHT LOSS AID - A health management system provides instantaneous feedback as to the relationship of food items and exercise to one's fitness level, including one's weight. The health management system does not require the user to count calories, either on the intake or expenditure side of the weight loss paradigm. Rather, the health management system may use icons and graphic displays, without units, to provide a user-friendly interface. The health management system can integrate weight, food intake and activity and can learn the individual's unique response to each element to predict the direction of weight gain or loss.08-30-2012
20120221494REGULAR EXPRESSION PATTERN MATCHING USING KEYWORD GRAPHS - Expanding a regular expression set into an expanded expression set that recognizes a same language as the regular expression set and includes more expressions than the regular expression set, with less operators per expression includes: logically connecting the expressions in the regular expression set; parsing the expanded expression set; transforming the parsed expanded expression set into a Glushkov automata; transforming the Glushkov automata into a modified deterministic finite automaton in order to maintain fundamental graph properties; combining the modified DFA into a keyword graph using a combining algorithm that preserves the fundamental graph properties; and computing an Aho-Corasick fail function for the keyword graph using a modified algorithm to produce a modified Aho-Corasick graph with a goto and a fail function and added information per state.08-30-2012
20120130925DECOMPOSABLE RANKING FOR EFFICIENT PRECOMPUTING - Methods and computer storage media are provided for generating an algorithm used to provide preliminary rankings to candidate documents. A final ranking function that provides final rankings for documents is analyzed to identify potential preliminary ranking features, such as static ranking features that are query independent and dynamic atom-isolated components that are related to a single atom. Preliminary ranking features are selected from the potential preliminary ranking features based on many factors. Using these selected features, an algorithm is generated to provide a preliminary ranking to the candidate documents before the most relevant documents are passed to the final ranking stage.05-24-2012
20120130927Shipping System and Method with Taxonomic Tariff Harmonization - A system, method and computer-readable medium for providing a harmonized classification code for a good based on input including a database adapted to store content including a harmonized tariff classification code module for storing a data structure representing a harmonized classification code tree, the harmonized classification code tree having one or more harmonized classification codes in which the good can be classified, a keywords module for associating and storing keyword data related to the good with one of the harmonized classification codes; and a learning module for learning keywords from the input and associating the learned key words with the one harmonized classification code for the good.05-24-2012
20120130926NETWORK CLASSIFICATION LINK SYSTEM, METHOD, AND COMPUTER RECORDING MEDIUM - A network classification link system, method, and computer recording medium are presented. The system includes: an operation interface, for providing an execution item menu and a network sheet; an input unit, for obtaining a first operation signal and a second operation signal to select an operating item and a network connection option corresponding to the first operation signal and the second operation signal respectively; a storage unit, for storing a network connection class, the operating item, and the network connection option; and a processing unit, for performing learning training through the operating item and the network connection option, generating a network classification link model according to the network connection class, the operating item, and the network connection, and judging a use weight of the network classification link model to establish an automatic network link mode.05-24-2012
20120166370SMART ATTRIBUTE CLASSIFICATION (SAC) FOR ONLINE REVIEWS - Techniques for identifying attributes in a sentence and determining a number of attributes to be associated with the sentence is described.06-28-2012
20120166369METHOD FOR DETERMINING HARMFUL MULTIMEDIA CONTENT USING MULTIMEDIA CONTENT PLAYBACK CHARACTERISTICS - A method for determining harmful multimedia content by using multimedia content playback characteristics includes: determining a local harmfulness of each basic unit section of multimedia content to generate a local determination result; and generating global determination results to complement an error of the local determination result based on the multimedia content playback characteristics. The global determination results are generated by using a continuous determination value which has a meaning of harmful or harmless and is updated depending on each local determination result and the number of continuous determination results is counted or initialized depending on continuity of the local determination results.06-28-2012
20120166368APPARATUS FOR GENERATING A PROBABILITY GRAPH MODEL USING A COMBINATION OF VARIABLES AND METHOD FOR DETERMINING A COMBINATION OF VARIABLES - An apparatus and method for generating a probability graph model are provided. When generating a probability graph model using variable combinations, a variable combination that has a small amount of information may not generated, thereby reducing the amount of computation. The apparatus may acquire independent variables including a plurality of input variables corresponding to context information and an output variable corresponding to an inference result, and may determine a variable combination that is to be generated, based on the amount of information of each of variable combinations with respect to the output value, in which the variable combination is defined based on combining of the input variables.06-28-2012
20120166367LOCATING A USER BASED ON AGGREGATED TWEET CONTENT ASSOCIATED WITH A LOCATION - A user submitting a query from a computer at an unknown location is located using a language model. The language model is derived from an aggregation of tweets that were sent from known locations.06-28-2012
20120166366HIERARCHICAL CLASSIFICATION SYSTEM - The claimed subject matter provides a method for hierarchical classification. The method includes receiving a hierarchical structure with a first level comprising a parent node and a sibling node. The structure also includes a second level comprising two child nodes. The method further includes receiving training examples. Each training example may be associated with a class of the parent node, the sibling node, or the two child nodes. The method also includes generating a first classifier for the first level. The first classifier includes a first hyperplane distinguishing the parent and sibling nodes. A first vector is normal to the first hyperplane. Additionally, the method includes generating a second classifier for the second level. The second classifier includes a second hyperplane distinguishing the two child nodes. A second vector is normal to the second hyperplane. An orthogonality of the second vector in relation to the first vector is maximized.06-28-2012
20100241598METHOD, PROGRAM, AND APPARATUS FOR GENERATING TWO-CLASS CLASSIFICATION/PREDICTION MODEL - A two-class classification/prediction model is generated in a simple operation by performing two-class classification with a classification rate substantially close to 100%. The two-class classification/prediction model is generated by a) obtaining a discriminant function for classifying a training sample set into two predetermined classes on the basis of an explanatory variable generated for each sample contained in the training sample set, b) calculating a discriminant score for each training sample by using the obtained discriminant function, c) determining, based on the calculated discriminant score, whether the training sample is correctly classified or not, d) determining a misclassified-sample region based on maximum and minimum discriminant scores taken from among misclassified samples in the training sample set, e) constructing a new training sample set by extracting the training samples contained in the misclassified-sample region, and f) repeating a) to e) for the new training sample set.09-23-2010
20100223212TASK-RELATED ELECTRONIC COACHING - Providing for task-related electronic feedback based on user interaction with a communication network is described herein. By way of example, user interactions the network or a network interface can be monitored to identify user activities performed in conjunction with a task. A rating for performance of the task can be obtained via comparison of user activities with benchmark performance activities. Based on the rating and user-benchmark comparison, inefficiencies can be identified, along with corrective actions for such activities. The corrective actions can then be output to coach the user on techniques for improving performance of the task. Accordingly, by employing corrective feedback based on monitored user activity, personal training can be automated, potentially reducing time and cost of such training.09-02-2010
20100205125IDENTIFYING INVENTION FEATURE PERMUTATIONS FOR A REASONABLE NUMBER OF PATENT APPLICATION CLAIMS - Permutations of features of an invention are ranked in accordance with factors such as importance and specificity to identify a reasonable number of (i.e. 20 or fewer) permutations as candidates for structuring a corresponding number of claims for a patent application. The identified permutations desirably include permutations corresponding to claims of broad scope, claims of narrow scope, and claims of intermediate scope; and exclude illogical or impractical permutations of features.08-12-2010
20100205124SUPPORT VECTOR MACHINE-BASED METHOD FOR ANALYSIS OF SPECTRAL DATA - Support vector machines are used to classify data contained within a structured dataset such as a plurality of signals generated by a spectral analyzer. The signals are pre-processed to ensure alignment of peaks across the spectra. Similarity measures are constructed to provide a basis for comparison of pairs of samples of the signal. A support vector machine is trained to discriminate between different classes of the samples. to identify the most predictive features within the spectra. In a preferred embodiment feature selection is performed to reduce the number of features that must be considered.08-12-2010
20100205123SYSTEMS AND METHODS FOR IDENTIFYING UNWANTED OR HARMFUL ELECTRONIC TEXT - The present invention relates to systems and methods for identifying and removing unwanted or harmful electronic text (e.g., spam). In particular, the present invention provides systems and methods utilizing inexact string matching methods and machine learning and non-learning methods for identifying and removing unwanted or harmful electronic text.08-12-2010
20100205122METHODS AND SYSTEMS OF ADAPTIVE COALITION OF COGNITIVE AGENTS - Coalitions from interactions and adaptations of cognitive map agents are evolved using an algorithm. A population of agents are seeded with cognitive map variants characterizing different cultures or different affiliations. The algorithm evolves this population by modifying the cognitive maps using a modified Particle Swarm Optimization algorithm. The modifications include modification to weights of the cognitive map, and the structure of the cognitive map of the global best (gbest) in the neighborhood is imitated according to a weighted random selection, based on the commonality of the node characteristic in the neighborhood. The end results indicate whether a coalition is possible and what cognitive maps emerge. These results are visualized on a 2D grid and measured with a clustering metric.08-12-2010
20100205121ASSOCIATIVE MEMORY LEARNING AGENT FOR ANALYSIS OF MANUFACTURING NON-CONFORMANCE APPLICATIONS - A system for assisting a user in determining a cause of a manufacturing non-conformance situation in a manufacturing application. The system may include an associative memory subsystem that is populated with a plurality of entity types, with each entity type including at least one entity, to form an associative memory. A user input device enables a user to input manufacturing non-conformance information into the associative memory subsystem that causes the associative memory subsystem to perform an initial search. The initial search generates a plurality of the entities that has a primary relevance useful for investigating the manufacturing non-conformance situation. An output device is responsive to the associative memory subsystem presents the plurality of entities found during the initial search to the user.08-12-2010
20100205120PLATFORM FOR LEARNING BASED RECOGNITION RESEARCH - A method for researching and developing a recognition model in a computing environment, including gathering one or more data samples from one or more users in the computing environment into a training data set used for creating the recognition model, receiving one or more training parameters defining a feature extraction algorithm configured to analyze one or more features of the training data set, a classifier algorithm configured to associate the features to a template set, a selection of a subset of the training data set, a type of the data samples, or combinations thereof, creating the recognition model based on the training parameters, and evaluating the recognition model.08-12-2010
20100070440SV REDUCTION METHOD FOR MULTI-CLASS SVM - An SV reduction method for multi-class SVMs is provided with which a number of SVs contained in the multi-class SVMs can be reduced without becoming trapped in a local minimum optimization solution and the reduction of the SVs can be performed at high precision and high speed. The method includes a step of selecting, from a plurality of initially present support vectors, support vector pairs z03-18-2010
20110184896METHOD FOR VISUALIZING FEATURE RANKING OF A SUBSET OF FEATURES FOR CLASSIFYING DATA USING A LEARNING MACHINE - A method for enhancing knowledge discovery from a dataset uses visualization of a subset features within a dataset that provide the best separation of the dataset into classes. One or more classifiers are trained using each subset of features and the success rate of the classifiers in accurately classifying the dataset is calculated. The success rate is converted into a ranking that is represented as a visually distinguishable characteristic. One or more tree structures may be displayed with a node representing each feature, and the visually distinguishable characteristic is used to indicate the scores for each feature subset. Connectors between the nodes may be used to indicate unconstrained and constrained feature sets. Nodes within a constrained path may be substituted for a feature within the preferred, unconstrained path if that feature is impractical to measure.07-28-2011
20110184895TRAFFIC OBJECT RECOGNITION SYSTEM, METHOD FOR RECOGNIZING A TRAFFIC OBJECT, AND METHOD FOR SETTING UP A TRAFFIC OBJECT RECOGNITION SYSTEM - A method for setting up a traffic object recognition system. A scene generator simulates three-dimensional simulations of various traffic situations which include at least one of the traffic objects. A projection unit generates signals which correspond to signals that the sensor would detect in a traffic situation simulated by the three-dimensional simulation. The signals are sent to the evaluation unit for recognizing traffic objects, and the pattern recognition is trained based on a deviation between the traffic objects simulated in the three-dimensional simulations of traffic situations and the traffic objects recognized therein.07-28-2011
20110184894GENERATING A SET OF ATOMS - An automated method comprises receiving training data representing an initial data set including text representing at least one concept embodied by the data set, using the training data in order to generate a set of atoms, each atom comprising at least one word that represents one or more concepts of the initial data set, wherein generating a set of atoms comprises minimising a cost function using an iterative process to identify one or more atoms.07-28-2011
20100179930Method and System for Developing Predictions from Disparate Data Sources Using Intelligent Processing - Provided herein is a platform for prediction based on extraction of features and observations collected from a large number of disparate data sources that uses machine learning to reinforce quality of collection, prediction and action based on those predictions.07-15-2010
20080235165Weak hypothesis generation apparatus and method, learning aparatus and method, detection apparatus and method, facial expression learning apparatus and method, facial enpression recognition apparatus and method, and robot apparatus - A facial expression recognition system that uses a face detection apparatus realizing efficient learning and high-speed detection processing based on ensemble learning when detecting an area representing a detection target and that is robust against shifts of face position included in images and capable of highly accurate expression recognition, and a learning method for the system, are provided. When learning data to be used by the face detection apparatus by Adaboost, processing to select high-performance weak hypotheses from all weak hypotheses, then generate new weak hypotheses from these high-performance weak hypotheses on the basis of statistical characteristics, and select one weak hypothesis having the highest discrimination performance from these weak hypotheses, is repeated to sequentially generate a weak hypothesis, and a final hypothesis is thus acquired. In detection, using an abort threshold value that has been learned in advance, whether provided data can be obviously judged as a non-face is determined every time one weak hypothesis outputs the result of discrimination. If it can be judged so, processing is aborted. A predetermined Gabor filter is selected from the detected face image by an Adaboost technique, and a support vector for only a feature quantity extracted by the selected filter is learned, thus performing expression recognition.09-25-2008
20100174671SYSTEM AND METHOD FOR CONCURRENTLY CONDUCTING CAUSE-AND-EFFECT EXPERIMENTS ON CONTENT EFFECTIVENESS AND ADJUSTING CONTENT DISTRIBUTION TO OPTIMIZE BUSINESS OBJECTIVES - The present invention is directed to systems, articles, and computer-implemented methods for assessing effectiveness of communication content and optimizing content distribution to enhance business objectives. Embodiments of the present invention are directed to computer-implemented methods for a computer-implemented method, comprising conducting an experiment using experimental content to determine effectiveness of communication content and executing, while conducting the experiment, a machine learning routine (MLR) using MLR content to enhance an effectiveness metric.07-08-2010
20100174670DATA CLASSIFICATION AND HIERARCHICAL CLUSTERING - Apparatus, systems, and methods can operate to provide efficient data clustering, data classification, and data compression. A method comprises training set of training instances can be processed to select a subset of size-1 patterns, initialize a weight of each size-1 pattern, include the size-1 patterns in classes in a model associated with the training set, and then include a set of top-k size-2 patterns in a way that provides an effective balance between local, class, and global significance patterns. A method comprises processing a dataset to compute an overall significance value of each size-2 pattern in each instance in the dataset, sort the size-2 patterns, and select the top-k size-2 patterns to be represented in clusters, which can be refined into a clustered hierarchy. A method comprises creating an uncompressed bitmap, reordering the bitmap, and compressing the bitmap. Additional apparatus, systems, and methods are disclosed.07-08-2010
20100049675Recovery of 3D Human Pose by Jointly Learning Metrics and Mixtures of Experts - Systems and methods are disclosed for determining human pose by generating an Appearance and Position Context (APC) local descriptor that achieves selectivity and invariance while requiring no background subtraction; jointly learning visual words and pose regressors in a supervised manner; and estimating the human pose.02-25-2010
20100049674GENERIC CLASSIFICATION SYSTEM - A classification system including a training device and one or more classification device for classifying one or more vectors other than training vectors. The training device is for selecting which training classification algorithms best classifies a set of training vectors, and for finding a set of values, of parameters of a generic classification algorithm, that enable the generic classification algorithm to substantially emulate the selected training classification algorithm.02-25-2010
20100049676Arrangement and Method for Network Management - The present invention relates to an arrangement for network management and adapted to be provided in or associated with a network node to be managed. It comprises, or is in communication with, modeling means adapted to, using substantially non-formal descriptions, model network domain and behavior using formal ontologies comprising inference capabilities by means of an inference engine, thus providing a formal ontology model describing domain and behavior. It also comprises annotating means adapted to add semantic information to the formal domain and behavior ontology model, generating means adapted to, using said formal ontology model and said inference engine, elaborate an algorithm adapted to generate and update a probabilistic causal network graph structure representing the domain and its behavior.02-25-2010
20120173467CONSTRUCTION OF AN AGENT THAT UTILIZES AS-NEEDED CANONICAL RULES - A method for constructing an agent that utilizes an as-needed canonical rule set in a first execution environment comprising requesting the as-needed rule set for the agent, supplying the agent with the as-needed rule set and requesting compilation of the as-needed rule set.07-05-2012
20120173466AUTOMATIC ANALYSIS OF LOG ENTRIES THROUGH USE OF CLUSTERING - A set of log entries is automatically inspected to determine a bug. A training set is utilized to determine clustering of log identifications. Log entries are examined in real-time or retroactively and matched to clusters. Timeframe may also be matched to a cluster based on log entries associated with the timeframe. Error indications may be outputted to a user of the system in respect to a log entry or a timeframe.07-05-2012
20120173465Automatic Variable Creation For Adaptive Analytical Models - A system and method for automated variable creation for adaptive fraud analytics are disclosed. A data structure for creation of rules is generated. The data structure represents nodes and associations between nodes from inputs for fraud/non-fraud conditions, and is generated from fraud and non-fraud data collected in an adaptive modeling process from past transactions. All unique paths between nodes of the data structure are determined to define a rule for each path. Each rule is then converted to a binary indicator variable to generate a set of binary indicator variables, and one or more complex variables is derived from the set of binary indicator variables. The one or more binary indicator variables and one or more complex variables can be provided to an adaptive scoring engine to score new transactions or to predict future behaviors.07-05-2012
20120254077Data Driven Frequency Mapping for Kernels Used in Support Vector Machines - Frequency features to be used for binary classification of data using a linear classifier are selected by determining a set of hypotheses in a d-dimensional space using d-dimensional labeled training data. A mapping function is constructed for each hypothesis. The mapping functions are applied to the training data to generate frequency features, and a subset of the frequency are selecting iteratively. The linear function is then trained using the subset of frequency features and labels of the training data.10-04-2012
20120215727AUTOMATIC DATA CLEANING FOR MACHINE LEARNING CLASSIFIERS - Systems and techniques for improving the training of machine learning classifiers are disclosed. A classifier is trained using a set of validated documents that are accurately associated with a set of class labels. A subset of non-validated documents is also identified and is used to further train and improve accuracy of the classifier.08-23-2012
20120259802ACTIVE LEARNING OF RECORD MATCHING PACKAGES - An active learning record matching system and method for producing a record matching package that is used to identify pairs of duplicate records. Embodiments of the system and method allow a precision threshold to be specified and then generate a learned record matching package having precision greater than this threshold and a recall close to the best possible recall. Embodiments of the system and method use a blocking technique to restrict the space of record matching packages considered and scale to large inputs. The learning method considers several record matching packages, estimates the precision and recall of the packages, and identifies the package with maximum recall having precision greater than equal to the given precision threshold. A human domain expert labels a sample of record pairs in the output of the package as matches or non-matches and this labeling is used to estimate the precision of the package.10-11-2012
20120259801TRANSFER OF LEARNING FOR QUERY CLASSIFICATION - Transfer of learning trains a new domain for the classification of search queries according to different tasks, as well as the generation of a corresponding domain-specific query classifier that may be used to classify the search queries according to the different tasks in the new domain. The transfer of learning may include preparing a new domain to receive classification knowledge from one or more source domains by populating the new domain with preliminary query patterns extracted for a search engine log. The transfer of learning may further include preparing the classification knowledge in each source domain for transfer to the new domain. The classification knowledge in each source domain may then be transferred to the new domain.10-11-2012
20100299290Web Query Classification - A query phrase may be automatically classified to one or more topics of interest (e.g., categories) to assist in routing the query phrase to one or more appropriate backend databases. A selectional preference query classification technique may be used to classify the query phrase based on a comparison between the query phrase and patterns of query phrases. Additionally, or alternatively, a combination of query classification techniques may be used to classify the query phrase. Topical classification of a query phrase also may be used to assist a search system in delivering auxiliary information to a user who entered the query phrase. Advertisements, for instance, may be tailored based on classification rather than query keywords.11-25-2010
20100299289SYSTEM AND METHOD FOR OBTAINING INFORMATION ABOUT BIOLOGICAL NETWORKS USING A LOGIC BASED APPROACH - A system and method of obtaining information concerning the structure-function relationship of biological networks can be studied holistically through the ensemble characterization of all the networks that realize a given biological function. A logic-based approach enables significant advances in computability and concept development (minimality and reducibility). The approach is applied to a biologically relevant trajectory and reveals some interesting properties. By using the approach, a cell cycle network is decomposed into three components with the functioning of each component explained.11-25-2010
20100299288RULE-BASED VOCABULARY ASSIGNMENT OF TERMS TO CONCEPTS - Methods and systems are described that involve rule-based vocabulary assignment of terms to concepts. Instead of assigning individual terms to each concept in a conceptualization of a domain, such as taxonomy, ontology, and so on, production rules are defined and assigned to each concept. The production rules produce at least one term to name a concept by referring to semantically related concepts to this concept. The production rules may include context information specifying the context where a given rule is valid. The methods and systems can be used to improve search capabilities for entities by enabling easier annotation of large conceptualizations. Further, the methods and systems can improve user experience by allowing context specific naming of entities.11-25-2010
20100049677SEQUENCE LEARNING IN A HIERARCHICAL TEMPORAL MEMORY BASED SYSTEM - A hierarchy of computing modules is configured to learn a cause of input data sensed over space and time, and is further configured to determine a cause of novel sensed input data dependent on the learned cause. At least one of the computing modules has a sequence learner module configured to associate sequences of input data received by the computing module to a set of causes previously learned in the hierarchy.02-25-2010
20120179634SYSTEM AND METHODS FOR FINDING HIDDEN TOPICS OF DOCUMENTS AND PREFERENCE RANKING DOCUMENTS - Systems and methods are disclosed to perform preference learning on a set of documents includes receiving raw input features from the set of documents stored on a data storage device; generating polynomial combinations from the raw input features; generating one or more parameters; applying the parameters to one or more classifiers to generate outputs; determining a loss function and parameter gradients and updating parameters determining one or more sparse regularizing terms and updating the parameters; and expressing that one document is preferred over another in a search query and retrieving one or more documents responsive to the search query.07-12-2012
20120123977INFORMATION PROCESSING APPARATUS, AND METHOD, INFORMATION PROCESSING SYSTEM, AND PROGRAM - Disclosed is an information processing apparatus including: a learning unit that learns user preference for each type in each category for classifying content items in a server; a selection unit that, based on type information indicating a recommendable type which is a type of content items recommendable by the server and a substitutable type which is a type that satisfies a predetermined condition out of the recommendable type, selects one or more recommendable types in a case where there is the recommendable type corresponding with user preference in the selected category, and selects one or more substitutable types in the selected category in a case where there is no recommendable type corresponding with user preference; and an obtaining unit that obtains a content of the selected type from the server.05-17-2012
20080270328Building and Using Intelligent Software Agents For Optimizing Oil And Gas Wells - A system and method for monitoring processes in the production of oil and gas uses intelligent software agents employing associative memory techniques that receive data from sensors in the production environment and from other sources and perform pattern matching operations to identify normal and abnormal behavior of the well production. The agents report the behaviors to human operators or other software systems. The abnormal behavior may consist of any behavior of the production processes that is other than the desired behavior of the well. The intelligent software agents are trained to identify both specific behaviors and behaviors that have never before been observed and recognized in the well.10-30-2008
20100010944MANAGING PERSONAL DIGITAL ASSETS OVER MULTIPLE DEVICES - In a first embodiment of the present invention, a method for managing digital assets of a user over multiple home network-enabled devices, the method comprising: receiving information, from a plurality of home network-enabled personal devices, regarding digital assets accessed by the personal devices, wherein the plurality of personal devices are owned or operated by the user and the information is automatically gathered by each personal device tracking its own usage; storing the information; and providing, to one of the plurality of personal devices, identifications of digital assets accessed by the personal devices by accessing the stored information.01-14-2010
20100010945METHOD AND SYSTEM FOR WEB RESOURCE LOCATION CLASSIFICATION AND DETECTION - A method and system for identifying locations associated with a web resource is provided. The location system identifies three different types of geographic locations: a provider location, a content location, and a serving location. A provider location identifies the geographic location of the entity that provides the web resource. A content location identifies the geographic location that is the subject of the web resource. A serving location identifies the geographic scope that the web page reaches. An application can select to use the type of location that is of particular interest.01-14-2010
20100010943LEARNING DEVICE, LEARNING METHOD, AND PROGRAM - A learning device includes: a plurality of learning modules, each of which performs update learning to update a plurality of model parameters of a pattern learning model that learns a pattern using input data; model parameter sharing means for causing two or more learning modules from among the plurality of learning modules to share the model parameters; module creating means for creating a new learning module corresponding to new learning data for learning the pattern when the new learning data are supplied as the input data; similarity evaluation means for evaluating similarities among the learning modules after the update learning is performed over all the learning modules including the new learning module; and module integrating means for determining whether to integrate the learning modules on the basis of the similarities among the learning modules and integrating the learning modules.01-14-2010
20100010942INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing device includes a first learning unit, a first error modeling unit, a first error generation unit, and a first estimation unit. The first learning unit learns a first estimation formula for estimating a first target variable of content on the basis of the feature quantity of the content. The first error modeling unit determines a first model of an error generated in the estimation of the first target variable based on the first estimation formula learned by the first learning unit. The first error generation unit generates, with the use of a random number, an error according to the first model determined by the first error modeling unit. The first estimation unit estimates the first target variable of the content by using the first estimation formula learned by the first learning unit and the random number generated by the first error generation unit.01-14-2010
20100010941Computer method and apparatus for classifying objects - A computer classification method and apparatus employs statistical analysis of known objects in the class of interest. For each known object in the class, a respective vector of q bits is formed. Each bit indicates presence or absence of an activity or physical property in the object. The probability that a bit is equal to 1 in the class is then applied to vector representations of test objects and determines probability of the test object belonging to the class.01-14-2010
20100010940Method for probabilistic information fusion to filter multi-lingual, semi-structured and multimedia Electronic Content - The invention belongs to the field of information system technology and more specifically in the area of electronic content management. The invention concerns method producing filtering systems of electronic documents that contain text in different languages, e.g. English, French, etc., as well as multimedia elements, e.g. digital images and/or digital video and/or digital excerpts of audio/speech. These documents can be semi-structured, i.e., they can exhibit structural features that are not to be found in non-digital documents, e.g. hyperlinks, or not.01-14-2010
20120185418SYSTEM AND METHOD FOR DETECTING ABNORMAL AUDIO EVENTS - Techniques for detecting abnormal audio events in a given environment, including learning the modeling of the environment to be surveilled during which a database is created by extraction of acoustic parameters associated with audio streams picked up over a fixed time period and an unsupervised automatic segmentation of said streams, followed by grouping the segments in classes and a statistical modeling of the segment classes, a usage phase including analysis of an audio stream, with the extraction of the acoustic parameters, automatic segmentation of said analysed stream substantially identical to that used during the learning phase and determining a likelihood of each statistical model contained in the database for each of the segments of the analysed audio stream, resulting in a likelihood value which is compared to a threshold value to determine the presence or absence of audio anomalies in the analysed audio stream.07-19-2012
20120185419DETERMINING A DYNAMIC USER PROFILE INDICATIVE OF A USER BEHAVIOR CONTEXT WITH A MOBILE DEVICE - Methods, apparatuses and articles of manufacture for use in a mobile device to determine whether a dynamic user profile is to transition from a first state to a second state based, at least in part, on one or more sensed indicators. The dynamic user profile may be indicative of one or more current inferable user behavior contexts for a user co-located with the mobile device. The mobile device may transition a dynamic user profile from a first state to a second state, in response to a determination that the dynamic user profile is to transition from the first state to the second state, and operatively affect one or more functions performed, at least in part, by the mobile device based, at least in part, on the transition of the dynamic user profile to the second state.07-19-2012
20120185415SYSTEM AND METHOD FOR DOMAIN ADAPTION WITH PARTIAL OBSERVATION - System, method and computer program product provides a novel domain adaption/transfer learning approach applied to the problem of classifying abbreviated documents, e.g., short text messages, instant messages, tweets. The proposed method uses a large number of multi-labeled examples (source domain) to improve the learning on the partial observations (target domain). Specifically, a hidden, higher-level abstraction space is learned that is meaningful for the multi-labeled examples in the source domain. This is done by simultaneously minimizing the document reconstruction error and the error in a classification model learned in the hidden space using known labels from the source domain. The partial observations in the target space are then mapped to the same hidden space, and classified into the label space determined by the source domain. Exemplary results provided for a Twitter dataset demonstrate that the method identifies meaningful hidden topics and provides useful classifications of specific tweets.07-19-2012
20120185417APPARATUS AND METHOD FOR GENERATING ACTIVITY HISTORY - According to one embodiment, a context acquisition unit acquires a context of a user and a date when the context has occurred. A context storage unit stores the context and the date. An activity information storage unit stores activity information of the user and date information to schedule the activity information. A first assignment unit assigns, to a first date corresponding to the date information, the activity information or an activity label extracted from the activity information. A second assignment unit assigns, to a second date to which the activity information or the activity label is not assigned, an activity label by using the context of the second date and an activity label assignment rule previously trained.07-19-2012
20120185416LOAD ESTIMATION IN USER-BASED ENVIRONMENTS - Method, system, and computer program product for load estimation in a user-based environment. The method includes: inputting a set of time-dependent, raw operational indicators of the environment; creating a load function according to the specific needs of the environment; displaying an estimated load; receiving user feedback on the estimated load; and applying a dynamic learning mechanism to generated a user-tuned load function for estimating load on the environment. The dynamic learning mechanism may be an informative mechanism that supports backtracking to solve user-adaptability problems.07-19-2012
20090327173METHOD FOR PREDICTING CYCLE TIME - A method for predicting cycle time comprises the steps of: collecting a plurality of known sets of data; using a clustering method to classify the known sets of data into a plurality of clusters; using a decision tree method to build a classification rule of the clusters; building a prediction model of each cluster; preparing data predicted set of data; using the classification rule to determine that to which clusters the predicted set of data belongs; and using the prediction model of the cluster to estimate the objective cycle time of the predicted set of data. Therefore, engineers can beforehand know the cycle time that one lot of wafers spend in the forward fabrication process, which helps engineers to properly arrange the following fabrication process of the lot of wafer.12-31-2009
20090327172ADAPTIVE KNOWLEDGE-BASED REASONING IN AUTONOMIC COMPUTING SYSTEMS - A method, information processing system, and network select machine learning algorithms for managing autonomous operations of network elements. A state (12-31-2009
20120221497Regular Expression Processing Automaton - A method and corresponding apparatus are provided implementing a stage one of run time processing using Deterministic Finite Automata (DFA) and implementing a stage two of run time processing using Non-Deterministic Finite Automata (NFA) to find the existence of a pattern in a payload, such as the payload portion of an Internet Protocol (IP) datagram, or an input stream.08-30-2012
20090018983METHOD AND SYSTEM FOR DETECTING ANOMALOUS PROCESS BEHAVIOR - A method for learning a process behavior model based on a process past instances and on one or more process attributes, and a method for detecting an anomalous process using the corresponding process behavior model.01-15-2009
20090018984System and method for dynamic knowledge construction - A system and method responsive to input stimuli is provided by incorporating a computer software program, hardware processing engine, or a specialized ASIC chip processor apparatus to capture concurrent inputs that are responsive to training stimulation, store a model representing a synthesis of the captured inputs, and use the stored model to generate outputs in response to real-world stimulation. Human user forced-choice approval/disapproval generated descriptions and decisions may be dynamically mapped with conventionally presented information and sensor and control data. The model mapping is stored into and out of a conventional mass storage device, such as is used in a relational database for use in generating a response to the stimuli. By accessing commonly stored mappings, the system can be incorporated into a mixture of multiple domains and disciplines of users and can create a common understanding of knowledge and design concept contained within it through mutual interaction, and subsequent automatic modifications to a common relational database. The system and method is applicable to conventional storage and presentation devices, making it easily incorporated into a variety of commercial products, utilizing current commercial human-machine interfaces (e.g. Human-Machine Interface graphical user interface, or Graphical User Interface) and current mass storage devices. The system uses N-dimensional descriptions of observations and concepts in an infinitely expandable space, embracing elements of human thought. This allows the user to tailor this system to control operation of automated devices and appliances to reflect the individual's wishes and desires as a dynamic representation and mapping of user descriptions and decisions with information, sensor data, and device controls.01-15-2009
20120191630Updateable Predictive Analytical Modeling - Methods, systems, and apparatus, including computer programs encoded on one or more computer storage devices, for training and retraining predictive models. A series of training data sets for predictive modeling can be received, e.g., over a network from a client computing system. The training data included in the training data sets is different from initial training data that was used with multiple training functions to train multiple trained predictive models stored in a predictive model repository. The series of training data sets are used with multiple trained updateable predictive models obtained from the predictive model repository and multiple training functions to generate multiple retrained predictive models. An effectiveness score is generated for each of the retrained predictive models. A first trained predictive model is selected from among the trained predictive models included in the predictive model repository and the retrained predictive models based on their respective effectiveness scores.07-26-2012
20090018981LEARNING CLASSIFIERS USING COMBINED BOOSTING AND WEIGHT TRIMMING - A “Classifier Trainer” trains a combination classifier for detecting specific objects in signals (e.g., faces in images, words in speech, patterns in signals, etc.). In one embodiment “multiple instance pruning” (MIP) is introduced for training weak classifiers or “features” of the combination classifier. Specifically, a trained combination classifier and associated final threshold for setting false positive/negative operating points are combined with learned intermediate rejection thresholds to construct the combination classifier. Rejection thresholds are learned using a pruning process which ensures that objects detected by the original combination classifier are also detected by the combination classifier, thereby guaranteeing the same detection rate on the training set after pruning. The only parameter required throughout training is a target detection rate for the final cascade system. In additional embodiments, combination classifiers are trained using various combinations of weight trimming, bootstrapping, and a weak classifier termed a “fat stump” classifier.01-15-2009
20090018980MULTIPLE-INSTANCE PRUNING FOR LEARNING EFFICIENT CASCADE DETECTORS - A “Classifier Trainer” trains a combination classifier for detecting specific objects in signals (e.g., faces in images, words in speech, patterns in signals, etc.). In one embodiment “multiple instance pruning” (MIP) is introduced for training weak classifiers or “features” of the combination classifier. Specifically, a trained combination classifier and associated final threshold for setting false positive/negative operating points are combined with learned intermediate rejection thresholds to construct the combination classifier. Rejection thresholds are learned using a pruning process which ensures that objects detected by the original combination classifier are also detected by the combination classifier, thereby guaranteeing the same detection rate on the training set after pruning. The only parameter required throughout training is a target detection rate for the final cascade system. In additional embodiments, combination classifiers are trained using various combinations of weight trimming, bootstrapping, and a weak classifier termed a “fat stump” classifier.01-15-2009
20120265717LEARNING SITUATIONS VIA PATTERN MATCHING - Example methods, apparatuses, or articles of manufacture are disclosed herein that may be utilized, in whole or in part, to facilitate or support one or more operations or techniques for machine learning of situations via pattern matching or recognition.10-18-2012
20120233097Multiple Hypothesis Tracking - Embodiments described herein are directed to multiple hypothesis systems and methods for tracking observations that are domain agnostic and involves determining the probability that a given set of observations (i.e., a track) corresponds to a particular target, object or linked set of events. One embodiment described herein relates to cyber security tracking methods and systems.09-13-2012
20120233096OPTIMIZING AN INDEX OF WEB DOCUMENTS - Historical usage data related to user queries and training properties for a plurality of web pages is received and utilized to train a mathematical model to predict the likelihood of retrieval of a web page during a web search. Properties are extracted from the plurality of web pages in the index and the mathematical model is applied to the properties for each web page to calculate a sortrank value. The index is reordered based on the sortrank value such that the web pages most likely to be retrieved by a user submitting a search query appear first in the index. After a search query is received from a user the index is traversed in an order determined by the sortrank value. Responsive web pages are presented to the user in an order determined by a search engine ranking algorithm.09-13-2012
20120233101CLUSTER ANALYSIS SYSTEM AND METHOD TO IMPROVE SORTING PERFORMANCE - A method for classifying an unknown part includes acquiring a broadband frequency response for a plurality of parts in a training set of parts, the training set of parts including a plurality of non-flawed parts and a plurality of flawed parts, performing a statistical analysis on the broadband frequency responses to form a plurality of part subsets, the plurality of part subsets including at least one subset of non-flawed parts and at least one subset of flawed parts, and utilizing the plurality of part subsets to form a blended subset of parts, the blended subset of parts being used to classify an unknown part as either a defective part or a non-defective part. A tool for implementing the method is also described.09-13-2012
20120233100Active Learing Decision Engines - Systems and methods for active learning decision engines in accordance with embodiments of the invention are disclosed. In one embodiment of the invention, an active learning decision engine includes equivalence class storage, hypotheses storage, edge storage, test storage, where tests are related to hypotheses, observation storage; and a processor, where the processor is configured to determine a plurality of equivalence classes containing one or more hypotheses, determine a set of edges utilizing tests, where the edges in the set of edges span hypotheses in distinct equivalence classes, determine weights for the determined edges, select a test based on the determined weights, perform the selected test and observe the results of the performed test, remove edges from the set of edges utilizing the observed results, and select a hypothesis from the one or more hypotheses using the set of edges.09-13-2012
20120233099OPTIMIZATION PROBLEM SOLVING - One method includes assigning one of a number of predefined values to each of a number of shadow prices of the system, distributing the assigned predefined shadow price values to a number of sub-problems, wherein each sub-problem is associated with one of a number of subsystems of the system, performing an analysis, including: determining a parametric solution and a region of validity for each of the number of sub-problems, determining an intersection of the regions of validity of all the parametric solutions, determining whether the optimization problem is solved from the parametric solutions, determining one or more shadow price updates based on the parametric solutions, and distributing the updated shadow prices to sub-problems having a region of validity that does not include the updated shadow prices, and repeating the analysis using the updated shadow prices until the optimization problem is solved from the parametric solutions of the number of sub-problems.09-13-2012
20080301072ROBOT SIMULATION APPARATUS - A robot simulation apparatus including: a display section which displays models of at least a conveyance apparatus, an object, and a robot respectively laid out at predetermined positions; a movement condition designating section which designates a direction and a speed of movement of the object; a imaging condition designating section which designates a relative position of the camera with respect to the object and imaging condition in order to obtain a still image of the object located within an imaging area; a teaching model storage section which stores a teaching model of the object to be compared with the still image obtained with the camera; a grasping position calculating section which calculates a grasping position of the object to be grasped by the robot based on a position and an attitude of the object obtained by comparing the still image with the teaching model, and on the direction and the speed of movement of the object; and a teaching position setting section which sets a teaching position for said robot based on the grasping position.12-04-2008
20080301071Support Vector Inductive Logic Programming - A computer implemented method of particular, although not exclusive application to analysing a plurality of molecules which comprises computing a kernel function for each pair of the plurality of molecules, the kernel function being representative of the number of features present in both molecules of the pairs and using the kernel function in a kernel based learning algorithm to model the relationship between the features and a property of the molecules. The method is also applicable to predicting a numerical value representing a characteristic of a molecule and, more generally, modelling instances of data in a database. A particular, although again not exclusive application, is the prediction of toxicity of a molecule.12-04-2008
20080301070KERNELS AND METHODS FOR SELECTING KERNELS FOR USE IN LEARNING MACHINES - Learning machines, such as support vector machines, are used to analyze datasets to recognize patterns within the dataset using kernels that are selected according to the nature of the data to be analyzed. Where the datasets possesses structural characteristics, locational kernels can be utilized to provide measures of similarity among data points within the dataset. The locational kernels are then combined to generate a decision function, or kernel, that can be used to analyze the dataset. Where an invariance transformation or noise is present, tangent vectors are defined to identify relationships between the invariance or noise and the data points. A covariance matrix is formed using the tangent vectors, then used in generation of the kernel.12-04-2008
20110004578ACTIVE METRIC LEARNING DEVICE, ACTIVE METRIC LEARNING METHOD, AND PROGRAM - A metric application unit receives data under analysis having a plurality of attributes and a metric indicative of the distance between the data under analysis, calculates the distance between the data under analysis, and output and stores a data analysis result which is generated from an analysis on the data under analysis with a predetermined function, using the calculated distance between the data under analysis. A metric optimization unit generates side-information based on an indication of feedback information entered from the outside and including either similarities between the data under analysis, or the attributes, or a combination thereof, generates a metric which complies with a predetermined condition, based on the generated side information, and stores the generated metric in a metric learning result storage unit.01-06-2011
20110004577EMOTION MODEL, APPARATUS, AND METHOD FOR ADAPTIVELY MODIFYING PERSONALITY FEATURES OF EMOTION MODEL - Disclosed are an emotion model, and an apparatus and method for adaptively learning personality of the emotion model. The emotion model, which maintains personality information, creates emotion information according to the personality information and takes a predetermined behavior according to the emotion information. The personality information may change adaptively in correspondence to a user's response to the behavior performed by the emotion model. Accordingly, the emotion model may react adaptively to the user through interactions with the user.01-06-2011
20110004576Systems & methods for improving recognition results via user-augmentation of a database - A system improves recognition results. The system receives multimedia data and recognizes the multimedia data based on training data to generate documents. The system receives user augmentation relating to one of the documents or new documents from a user. The system supplements the training data with the user augmentation or new documents and retrains based on the supplemented training data.01-06-2011
20110004575System and Method for Controlling Power Consumption in a Computer System Based on User Satisfaction - Systems and methods for controlling power consumption in a computer system are disclosed. The computer system may be trained, for example, to determine relationship information between user satisfaction and discrete frequencies at which a processor of the computer system runs. The determined relationship can distinguish between different users and different interactive applications. A frequency may be selected from the discrete frequencies at which the processor of the computer system runs based on the determined relationship information for a particular user and a particular interactive application running on the processor of the computer system. The processor may be adapted to run at the selected frequency.01-06-2011
20110004574EXECUTION ALLOCATION COST ASSESSMENT FOR COMPUTING SYSTEMS AND ENVIRONMENTS INCLUDING ELASTIC COMPUTING SYSTEMS AND ENVIRONMENTS - Techniques for allocating individually executable portions of executable code for execution in an Elastic computing environment are disclosed. In an Elastic computing environment, scalable and dynamic external computing resources can be used in order to effectively extend the computing capabilities beyond that which can be provided by internal computing resources of a computing system or environment. Machine learning can be used to automatically determine whether to allocate each individual portion of executable code (e.g., a Weblet) for execution to either internal computing resources of a computing system (e.g., a computing device) or external resources of an dynamically scalable computing resource (e.g., a Cloud). By way of example, status and preference data can be used to train a supervised learning mechanism to allow a computing device to automatically allocate executable code to internal and external computing resources of an Elastic computing environment.01-06-2011
20110004573IDENTIFYING TRAINING DOCUMENTS FOR A CONTENT CLASSIFIER - Systems, methods and articles of manufacture are disclosed for identifying a training document for a content classifier. One or more thresholds may be defined for designating a document as a training document for a content classifier. A plurality of documents may be evaluated to compute a score for each respective document. The score may represent suitability of a document for training the content classifier with respect to a category. The score may be computed based on content of the plurality of documents, metadata of the plurality of documents, link structure of the plurality of documents, user feedback (e.g., user supplied document tags) received for the plurality of documents, and document metrics received for the plurality of documents. Based on the computed scores, a training document may be selected. The content classifier may be trained using the selected training document.01-06-2011
20100100510DYNAMIC DISCRETE DECISION SIMULATION SYSTEM - A system that enables dynamic discrete decision simulation is provided. Simulation has many advantages in modeling complex systems to facilitate decision making. The innovation discloses a system that integrates an agent-based discrete event simulator, a geographic information system, a rule base, and interactive databases in addition to interfaces and other supporting components. The modules can seamlessly communicate with each other by exchanging a progression of data, and by making a series of deductive decisions through embedded algorithms. The integrated system can be applied to disaster management planning and training.04-22-2010
20120239600METHOD FOR TRAINING AND USING A CLASSIFICATION MODEL WITH ASSOCIATION RULE MODELS - A classification model is trained and used for detecting patterns in input data. The training of the model includes retrieving a set of previously recorded input data containing a plurality of items associated with a plurality of entities and adding to each entity a known classification. Furthermore, training the model includes determining rules from the set of previously recorded input data and the known classification by associating the classification of each entity with the respective items of said entity. The training of the model further includes determining a set of rules which are applicable, aggregating the lift values of the rules determined for said entity, and predicting a classification based on the aggregated association values for each entity. The resulting aggregated lift value together with the respective entity and classification are used as input for a standard classification algorithm, where the result is a classification model.09-20-2012
20120239598Machine Learning Method to Identify Independent Tasks for Parallel Layout in Web Browsers - Methods and devices for accelerating web page rendering include processing web pages and gathering web page element information, performing machine learning analysis on the gathered web page element information to identify patterns in layout independence correlated to web page element information, and training a classifier to predict sub-tree independence based on element information in a web page script. The predicted sub-tree independence may be used to concurrently process portions of a web page to be rendered to reduce the time required to render the page. Sub-trees may be conditionally independent, in which case, the conditionally independent sub-trees may be made independent by speculating data to render the sub-trees independent, or by performing a task to obtain the certain information to render the sub-tree independent.09-20-2012
20110131161Methods and Systems for Selecting and Presenting Content on a First System Based on User Preferences Learned on a Second System - A method of selecting and presenting content on a first system based on user preferences learned on a second system is provided. The method includes receiving a user's input for identifying items of the second content system and, in response thereto, presenting a subset of items of the second content system and receiving the user's selection actions thereof. The method includes analyzing the selected items to learn the user's content preferences for the content of the second content system and determining a relationship between the content of the first and second content systems to determine preferences relevant to items of the first content system. The method includes, in response subsequent user input for items of the first content system, selecting and ordering a collection of items of the first content system based on the user's learned content preferences determined to be relevant to the items of the first content system.06-02-2011
20110131160Method and System for Generating A Linear Machine Learning Model for Predicting Online User Input Actions - A method of targeting receives several granular events and preprocesses the received granular events thereby generating preprocessed data to facilitate construction of a model based on the granular events. The method generates a predictive model by using the preprocessed data. The predictive model is for determining a likelihood of a user action. The method trains the predictive model. A system for targeting includes granular events, a preprocessor for receiving the granular events, a model generator, and a model. The preprocessor has one or more modules for at least one of pruning, aggregation, clustering, and/or filtering. The model generator is for constructing a model based on the granular events, and the model is for determining a likelihood of a user action. The system of some embodiments further includes several users, a selector for selecting a particular set of users from among the several users, a trained model, and a scoring module.06-02-2011
20110131159SYSTEMS AND METHODS FOR DETECTING THE PRESENCE OF A BIOLOGICAL STATUS USING CLUSTERING - A method for determining the presence of a biological entity. The method may include entering into a digital computer, at least a plurality of first input values associated with a first genetic element (e.g., mecA), a plurality of second input values associated with a second genetic element (femA), and a plurality of third input values associated with a third genetic element (e.g., orfX) associated with a plurality of samples. Each sample includes a first input value in the plurality of first input values, a second input value in the plurality of second input values, and a third input value in the plurality of third input values. The method also includes determining a threshold value associated with the third genetic element, separating the samples using the threshold value into a first set of samples and a second set of samples, clustering the first set of samples in a feature space defined by the first genetic element and the second genetic element, defining a first boundary space using the first set of samples, and defining a second boundary space using the second set of samples. The first and second boundary spaces differentiate a biological entity from other biological statuses. Other embodiments may also include the use of a genetic element such as SCCmec.06-02-2011
20110131158INFORMATION PROCESSING APPARATUS AND INFORMATION PROCESSING METHOD - An information processing apparatus that selects a plurality of feature amounts acquired by applying a filter to learning data and generates a discriminator based on the selected feature amounts includes a time specification unit configured to specify a calculation time required for acquiring a feature amount of a selection candidate by applying the filter to the selected feature amounts or the learning data, a precision specification unit configured to specify a precision of a discriminator generated based on the feature amount of the selection candidate and the selected feature amounts, a selection unit configured to select the feature amount of the selection candidate based on the calculation time and the precision, and a generation unit configured to generate the discriminator based on the selected feature amounts.06-02-2011
20110131157SYSTEM AND METHOD FOR PREDICTING CONTEXT-DEPENDENT TERM IMPORTANCE OF SEARCH QUERIES - An improved system and method for identifying context-dependent term importance of queries is provided. A query term importance model is learned using supervised learning of context-dependent term importance for queries and is then applied for advertisement prediction using term importance weights of query terms as query features. For instance, a query term importance model for query rewriting may predict rewritten queries that match a query with term importance weights assigned as query features. Or a query term importance model for advertisement prediction may predict relevant advertisements for a query with term importance weights assigned as query features. In an embodiment, a sponsored advertisement selection engine selects sponsored advertisements scored by a query term importance engine that applies a query term importance model using term importance weights as query features and inverse document frequency weights as advertisement features to assign a relevance score.06-02-2011
20110131156System, Method and Computer Program Product for Incremental Learning of System Log Formats - A computer program is disclosed including but not limited to instructions to input an initial description of a data format and a batch of data comprising data in a new data format not covered by the initial description, instructions to use the first description to parse the records in the data source, instructions to discard records in the input data that parse successfully, instructions to collect records that fail to parse, instructions to accumulate a quantity, M of records that fail to parse, instructions to return a modified description that extends the initial description to cover the new data, instructions to transform the first description, D into a second description D′ to accommodate differences between the input data format and the first description D by introducing options where a piece of data was missing in the input data and introducing unions where a new type of data was found in the input data; and instructions to use a non-incremental format inference system such as LEARNPADS to infer descriptions for the aggregated portions of input data that did not parse using the first description D.06-02-2011
20110131155Method for Determining Distributions of Unobserved Classes of a Classifier - A distribution of an unobserved class for a classifier with no known training data is learned by first determining, for each known class, known distribution using known training data. Sufficient statistics of the distribution of the unobserved class are determined from the known distributions and the training data associated with each known class. If the known training data and the known distributions are bounded, then update parameters of the distribution of the unobserved class from the sufficient statistics, else update the parameters from sufficient statistics and a priori probability distributions that specify the distributions of the parameters.06-02-2011
20120265716MACHINE LEARNING OF KNOWN OR UNKNOWN MOTION STATES WITH SENSOR FUSION - Example methods, apparatuses, or articles of manufacture are disclosed herein that may be utilized, in whole or in part, to facilitate or support one or more operations or techniques for machine learning of known or unknown motion states with sensor fusion.10-18-2012
20120323830METHOD FOR AUTOMATICALLY TEACHING PARAMETERS - The invention relates to a method for automatically teaching parameters to a tray sealer, for example, position values, acceleration values and/or distances.12-20-2012
20110047105Use of Machine Learning for Classification of Magneto Cardiograms - The use of machine learning for pattern recognition in magnetocardiography (MCG) that measures magnetic fields emitted by the electrophysiological activity of the heart is disclosed herein. Direct kernel methods are used to separate abnormal MCG heart patterns from normal ones. For unsupervised learning, Direct Kernel based Self-Organizing Maps are introduced. For supervised learning Direct Kernel Partial Least Squares and (Direct) Kernel Ridge Regression are used. These results are then compared with classical Support Vector Machines and Kernel Partial Least Squares. The hyper-parameters for these methods are tuned on a validation subset of the training data before testing. Also investigated is the most effective pre-processing, using local, vertical, horizontal and two-dimensional (global) Mahanalobis scaling, wavelet transforms, and variable selection by filtering.02-24-2011
20120271782Method and apparatus for event detection permitting per event adjustment of false alarm rate - Method and apparatus for object or event of interest detection which minimizes the level of false alarms and maximizes the level of detections as defined on a per event or object basis by the analyst. The invention allows for the minimization of false alarms for objects or events of interest which have a close resemblance to all other objects or events mapped to the same multidimensional feature space, and allows for the per event or per object adjustment on false alarms for objects or events of higher interest.10-25-2012
20120323827Generating Predictions From A Probabilistic Process Model - A method for predictive analytics in a semi-structured process including updating, iteratively, at least one probability of a probabilistic process model based on a completed task, wherein updating the at least one probability of the probabilistic process model includes receiving the probabilistic process model associated with a todo list including a plurality of tasks of the semi-structured process, defining a cost of each of the plurality of tasks, prioritizing the plurality of tasks according to the costs, and recommending a next task from the todo list according to a prioritization12-20-2012
20120323828FUNCTIONALITY FOR PERSONALIZING SEARCH RESULTS - A query processing system is described herein for personalizing results for a particular user. The query processing system operates by receiving a query from a particular user u who intends to find results that satisfy the query with respect to a topic T12-20-2012
20120323825SYSTEM AND METHODS FOR FINDING HIDDEN TOPICS OF DOCUMENTS AND PREFERENCE RANKING DOCUMENTS - Systems and methods are disclosed to perform preference learning on a set of documents includes receiving raw input features from the set of documents stored on a data storage device; generating polynomial combinations from the raw input features; generating one or more parameters; applying the parameters to one or more classifiers to generate outputs; determining a loss function and parameter gradients and updating parameters determining one or more sparse regularizing terms and updating the parameters; and expressing that one document is preferred over another in a search query and retrieving one or more documents responsive to the search query.12-20-2012
20120323829GRAPH-BASED CLASSIFICATION BASED ON FILE RELATIONSHIPS - A reliable automated malware classification approach with substantially low false positive rates is provided. Graph-based local and/or global file relationships are used to improve malware classification along with a feature selection algorithm. File relationships such as containing, creating, copying, downloading, modifying, etc. are used to assign malware probabilities and simultaneously reduce the false positive and false negative rates on executable files.12-20-2012
20110060705ALERT GENERATION SYSTEM AND METHOD03-10-2011
20120323826System and Method for Predicting Political Instability using Bayesian Networks - Disclosed is a system and method for predicting political instability. This instability is predicted for specific countries or geographic regions. In one embodiment, the prediction is carried out on a basis of a probabilistic model, such as a Bayesian-network. The model is comprised of various notes corresponding to dependent and independent variables. The independent variables, in turn, correspond to factors relating to historical political instability. The dependent variable corresponds to the prediction of instability. By populating the independent variables with current data, future political instability can be predicted.12-20-2012
20110238604COMPUTER-AIDED DIAGNOSTIC SYSTEMS AND METHODS FOR DETERMINING SKIN COMPOSITIONS BASED ON TRADITIONAL CHINESE MEDICINAL (TCM) PRINCIPLES - Computer-aided systems and methods are provided for determining the skin composition of a specific user according to Traditional Chinese Medicinal (TCM) principles by statistically analyzing biological and/or psychological information collected from such user, such as age, gender, bodily sensation, skin condition and complexion, sleep pattern, dietary habits, energy level, stress level, physical fitness and emotional wellness, so as to classify the skin composition of the user according to TCM principles but without employing a TCM practitioner. Preferably, the skin composition classification is indicative of Yin-Yang balance of the skin of the user or the lack thereof. The present systems and methods may further recommend to the user one or more topical skin care regimens and/or ingestible skin benefit products suitable for the skin composition of the specific user.09-29-2011
20110238603System and Method for Predicting Events Via Dynamic Ontologies - Disclosed is a system and method for determining the probability of an event occurring. The method involves developing models relating a number of factors and variables. The factors and variables can be unique to a specified field of endeavor, such as military security or epidemiology. The models can be ontological models. A rule set is then utilized to relate certain variables in the models to a specific event. The rule set can be embodied in a computer model, such as a Bayesian-Network. The system permits a user to query a knowledge store or database to acquire referent values for the rule set. Thereafter, the referent values are used to populate the rule set and compute the probability of the event occurring.09-29-2011
20100169249System and Method for Determining Semantically Related Terms Using an Active Learning Framework - Systems and methods for determining semantically related terms using an active learning framework such as Transductive Experimental Design are disclosed. Generally, to enhance a keyword suggestion tool, an active learning module trains a model to predict whether a term is relevant to a user. The model is then used to present the user with terms that have been determined to be relevant based on the model so that an online advertisement service provider may more efficiently provide a user with terms that are semantically related to a seed set.07-01-2010
20100169248CONTENT DIVISION POSITION DETERMINATION DEVICE, CONTENT VIEWING CONTROL DEVICE, AND PROGRAM - A content division position determination device 07-01-2010
20100169250METHODS AND SYSTEMS FOR TRANSDUCTIVE DATA CLASSIFICATION - A system, method, data processing apparatus, and article of manufacture are provided for classifying data. Labeled data points are received, each of the labeled data points having at least one label indicating whether the data point is a training example for data points for being included in a designated category or a training example for data points being excluded from a designated category; receiving unlabeled data points; receiving at least one predetermined cost factor of the labeled data points and unlabeled data points; training a transductive classifier using MED through iterative calculation using the at least one cost factor and the labeled data points and the unlabeled data points as training examples; applying the trained classifier to classify at least one of the unlabeled data points, the labeled data points, and input data points; and outputting a classification of the classified data points, or derivative thereof.07-01-2010
20100169245Statistical Machine Learning - Statistical machine learning, in which an input module receives user input that defines a hypothesis associated with a particular Output. The hypothesis defines one or more starting criteria that are proposed as being correlated with the particular output, and a recommendation engine initially provides recommendations that include the particular output based on the one or more starting criteria defined by the hypothesis. An experience analytics system receives feedback data related to whether the recommendations provided based on the one or more starting criteria defined by the hypothesis were successful and modifies the hypothesis based on the feedback data. Subsequent to the experience analytics system modifying the hypothesis, the recommendation engine provides recommendations that include the particular output based on the modified hypothesis.07-01-2010
20100169243METHOD AND SYSTEM FOR HYBRID TEXT CLASSIFICATION - A computer-implemented system and method for text classification is provided that applies a hybrid approach for text classification. The system and method includes a text pre-processor which prepares unclassified articles in a format which can be read by a two-stage classifier. The classifier employs a hybrid approach. A keyword-based model achieves machine-labelling of the articles. The machine-labelled articles are used to train a machine learning model. New articles can be applied against the trained model, and classified.07-01-2010
20100169246Multimodal system and input process method thereof - A multimodal system and an input processing method thereof are disclosed. The multimodal system includes a pre-constructed input combination constructing unit and an input combination selection unit for selecting an input combination corresponding to an input signal from a user or a sensor. The system performs learning for selecting an input combination from the pre-constructed input combinations. The system provides available input combinations due to this learning, resulting in high satisfaction with the processed result.07-01-2010
20100169244METHOD AND APPARATUS FOR USING A DISCRIMINATIVE CLASSIFIER FOR PROCESSING A QUERY - A method and apparatus for using a classifier for processing a query are disclosed. For example, the method receives a query from a user, and processes the query to locate one or more documents in accordance with a search engine having a discriminative classifier, wherein the discriminative classifier is trained with a plurality of artificial query examples. The method then presents a result of the processing to the user.07-01-2010
20100169247System and method for statistical measurment validation - An apparatus and method are disclosed for a measurement system that reports as a measurement result a confidence interval associated with a histogram bin into which a measurement value falls. The confidence interval is calculated from a subset of training values that also fall within the histogram bin. A training process may be performed in which a plurality of training values is obtained and a mean and standard deviation of the values determined. A plurality of histogram bins are defined from the mean and standard deviation and, for the subsets of training values that fall into each bin, confidence intervals calculated. A need to perform the training process may be determined from a plurality of measured values.07-01-2010
20100235307METHOD, SYSTEM, AND COMPUTER PROGRAM FOR USER-DRIVEN DYNAMIC GENERATION OF SEMANTIC NETWORKS AND MEDIA SYNTHESIS - This invention relates generally to classification systems. More particularly this invention relates to a system, method, and computer program to dynamically generate a domain of information synthesized by a classification system or semantic network. The invention discloses a method, system, and computer program providing a means by which an information store comprised of knowledge representations, such as a web site comprised of a plurality of web pages or a database comprised of a plurality of data instances, may be optimally organized and accessed based on relational links between ideas defined by one or more thoughts identified by an agent and one or more ideas embodied by the data instances. Such means is hereinafter referred to as a “thought network”.09-16-2010
20120278261DETERMINING THE IMPORTANCE OF DATA ITEMS AND THEIR CHARACTERISTICS USING CENTRALITY MEASURES - Computer-implemented methods, systems, and articles of manufacture for determining the importance of a data item. A method includes: (a) receiving a node graph; (b) approximating a number of neighbor nodes of a node; and (c) calculating a average shortest path length of the node to the remaining nodes using the approximation step, where this calculation demonstrates the importance of a data item represented by the node. Another method includes: (a) receiving a node graph; (b) building a decomposed line graph of the node graph; (c) calculating stationary probabilities of incident edges of a node graph node in the decomposed line graph, and (d) calculating a summation of the stationary probabilities of the incident edges associated with the node, where the summation demonstrates the importance of a data item represented by the node. Both methods have at least one step carried out using a computer device.11-01-2012
20120278262Suggesting Users for Interacting in Online Applications in a Social Networking Environment - Users of a social networking system are matched with other users of the social networking system based on the likelihood of both users' being interested in using a social application and the likelihood that they would want to interact with each other using the application. Social applications include social games and other applications associated with a social networking system in which users can interact with other users. The social networking system selects for a particular user other candidate users. This selection may be based on at least one of a predicted likelihood that a user would invite the candidate users to use the social application and/or a predicted likelihood that the selected candidate user would accept the user's invitation. The user may choose to act on the suggestion by inviting the other user and the invited user may accept, reject, or ignore the invitation.11-01-2012
20110258149RANKING SEARCH RESULTS USING CLICK-BASED DATA - Methods and computer-storage media having computer-executable instructions embodied thereon that facilitate generating a machine-learned model for ranking search results using click-based data are provided. Data is referenced from user queries, which may include search results generated by general search engines and vertical search engines. A training set is generated from the search results and click-based judgments are associated with the search results in the training set. Based on click-based judgments, identifiable features are determined from the search results in a training set. Based on determining identifiable features in a training set, a rule set is generated for ranking subsequent search results.10-20-2011
20110251981Enhanced Learning and Recognition Operations for Radial Basis Functions - Methods, apparatuses and systems directed to pattern identification and pattern recognition. In some particular implementations, the invention provides a flexible pattern recognition platform including pattern recognition engines that can be dynamically adjusted to implement specific pattern recognition configurations for individual pattern recognition applications. In some implementations, the present invention also provides for a partition configuration where knowledge elements can be grouped and pattern recognition operations can be individually configured and arranged to allow for multi-level pattern recognition schemes.10-13-2011
20110258150SYSTEMS AND METHODS FOR TRAINING DOCUMENT ANALYSIS SYSTEM FOR AUTOMATICALLY EXTRACTING DATA FROM DOCUMENTS - A method of training a document analysis system to extract data from documents is provided. The method includes: automatically analyzing images and text features extracted from a document to associate the document with a corresponding document category; comparing the extracted text features with a set of text features associated with corresponding category of the document, in which the set of text features includes a set of characters, words, and phrases; if the extracted features are found to consist of the characters, words, and phrases belonging to the set of text features associated with the corresponding document category, storing the extracted text features as the data contained in the corresponding document; and, if the extracted text features are found to include at least one text feature that does not belong to the set of text features associated with the corresponding document category, submitting the unrecognized text features to a training phase.10-20-2011
20120089543Regulated Data Analysis System - A data analysis system is invented to analysis business data. The analysis process is regulated to increase accuracy.04-12-2012
20120089542Consistency Maintenance of Distributed Graph Structures - The present disclosure is directed to systems and methods including retrieving a model including a plurality of objects and references between objects, receiving first user input indicating a set of first changes to the model, applying changes of the set of first changes to the model to provide a first modified model, receiving second user input indicating a set of second changes to the model, identifying a conflicting operation in the set of first changes to the set of second changes, applying one or more inverse operations to the first modified model to provide a second modified model, removing the conflicting operation from the set of first changes, defining a subset of first changes including the one or more changes after the conflicting operation, reconciling one or more changes to provide a reconciled subset of first changes, and defining an updated model.04-12-2012
20110276524Data Structures and Apparatuses for Representing Knowledge - Data structures and apparatuses to represent knowledge are disclosed. The processes can comprise labeling elements in a knowledge signature according to concepts in an ontology and populating the elements with confidence values. The data structures can comprise knowledge signatures stored on computer-readable media. The knowledge signatures comprise a matrix structure having elements labeled according to concepts in an ontology, wherein the value of the element represents a confidence that the concept is present in an information space. The apparatus can comprise a knowledge representation unit having at least one ontology stored on a computer-readable medium, at least one data-receiving device, and a processor configured to generate knowledge signatures by comparing datasets obtained by the data-receiving devices to the ontologies.11-10-2011
20110276523MEASURING DOCUMENT SIMILARITY BY INFERRING EVOLUTION OF DOCUMENTS THROUGH REUSE OF PASSAGE SEQUENCES - One embodiment of the present invention provides a system for estimating document similarity. During operation, the system selects a collection of documents which includes a first set of passages, constructs a passage-sequence model based on the first set of passages, receives a new document which includes a second set of passages, and determines a sequence of operations associated with the new document in relation to the collection of documents based on the constructed passage-sequence model.11-10-2011
20120095944Forward Feature Selection For Support Vector Machines - In one embodiment, the present invention includes a method for training a Support Vector Machine (SVM) on a subset of features (d′) of a feature set having (d) features of a plurality of training instances to obtain a weight per instance, approximating a quality for the d features of the feature set using the weight per instance, ranking the d features of the feature set based on the approximated quality, and selecting a subset (q) of the features of the feature set based on the ranked approximated quality. Other embodiments are described and claimed.04-19-2012
20120095943SYSTEM FOR TRAINING CLASSIFIERS IN MULTIPLE CATEGORIES THROUGH ACTIVE LEARNING - A system for training classifiers in multiple categories through an active learning system, including a computer having a memory and a processor, the processor programmed to: train an initial set of m binary one-versus-all classifiers, one for each category in a taxonomy, on a labeled dataset of examples stored in a database coupled with the computer; uniformly sample up to a predetermined large number of examples from a second, larger dataset of unlabeled examples stored in a database coupled with the computer; order the sampled unlabeled examples in order of informativeness for each classifier; determine a minimum subset of the unlabeled examples that are most informative for a maximum number of the classifiers to form an active set for learning; and use editorially-labeled versions of the examples of the active set to re-train the classifiers, thereby improving the accuracy of at least some of the classifiers.04-19-2012
20120330865SYSTEM AND METHOD FOR FORMULATING A PROBLEM - A method for formulating a problem using a computational system is provided. The method includes determining an initial problem statement that characterizes the problem and identifying a plurality of factors affecting the problem. The method also includes generating a plurality of hypotheses associated with the problem based upon the identified factors and updating the initial problem statement to an updated problem statement using the initial problem statement, identified factors and the plurality of hypotheses.12-27-2012
20120330868MATCHING APPARATUS AND NON-TRANSITORY COMPUTER READABLE STORAGE MEDIUM - The matching apparatus 12-27-2012
20120330867SYSTEMS AND METHODS FOR LARGE-SCALE RANDOMIZED OPTIMIZATION FOR PROBLEMS WITH DECOMPOSABLE LOSS FUNCTIONS - Systems and methods directed toward processing optimization problems using loss functions, wherein a loss function is decomposed into at least one stratum loss function, a loss is decreased for each stratum loss function to a predefined stratum loss threshold individually using gradient descent, and the overall loss is decreased to a predefined threshold for the loss function by appropriately ordering the processing of the strata and spending appropriate processing time in each stratum. Other embodiments and aspects are also described herein.12-27-2012
20120330866SYSTEM AND METHOD FOR DETERMINING AN OPTIMUM QC STRATEGY FOR IMMEDIATE RELEASE RESULTS - The present invention proposes a method for optimizing a quality control strategy for rapid release results. An embodiment of the invention includes generating a set of candidate quality control rules and for each candidate rule, computing a maximum number of patient specimens that can be tested between quality control events while keeping the expected number of correctable unacceptable results below a predetermined correctable maximum and keeping the expected number of final unacceptable results below a predetermined final maximum. Furthermore a quality control utilization rate can be computed based on the number of patient specimens tested between each quality control event and the number of reference samples tested at each quality control event. The candidate rule for which the best quality control utilization rate may be selected along with the corresponding number of patients to be tested between each quality control as the optimum quality control strategy.12-27-2012
20110307429AUTOMATED CLASSIFICATION ALGORITHM COMPRISING AT LEAST ONE INPUT-INVARIANT PART - A classification algorithm is separated into one or more input-invariant parts and one or more input-dependent classification parts. Classifiable electronic data is obtained via a communication network. Using the classification algorithm, classifications of a plurality of data elements in the classifiable data are identified, where the at least one classification part incorporates user input concerning classification of at least one data element of the plurality of data elements.12-15-2011
20110320389SYSTEMS AND METHODS FOR SAFETY AND BUSINESS PRODUCTIVITY - The present invention is a safety and business productivity system having the following components. One or more cameras capture video data having attribute data, the attribute data representing importance of the cameras. One or more video analytics devices process the video data from one or more of the cameras and detect primitive video events in the video data. A correlation engine correlates two or more primitive video events from the video analytics devices weighted by the attribute data of the cameras used to capture the video data. An alerting engine generates one or more alerts and performs one or more actions based on the correlation performed by the correlation engine.12-29-2011
20110320388System, Method and Computer Program for Pattern Based Intelligent Control, Monitoring and Automation - The present invention relates to control, monitoring, and automation. The present invention more specifically relates to pattern-based intelligent control, monitoring and automation. The invention performs pattern-based monitoring. It collects signal data from one or more signals. The signal data define signal data streams. It then transforms each of the signal data streams into trends. It also discovers patterns based on the trends within each signal data stream and/or across the signal data streams. The patterns are optionally used for diagnostics and root cause analysis, online plant monitoring and operation control, plant optimization, and other environments where a causal link or correlation may exist between related inputs, states and/or outputs.12-29-2011
20110320386EXTRAPOLATING EMPIRICAL MODELS FOR CONTROL, PREDICTION, AND OPTIMIZATION APPLICATIONS - The present disclosure provides novel techniques for defining empirical models having control, prediction, and optimization modalities. The empirical models may include neural networks and support vector machines. The empirical models may include asymptotic analysis as part of the model definition as allow the models to achieve enhanced results, including enhanced high-order behaviors. The high-order behaviors may exhibit gains that are non-zero trending, which may be useful for controller modalities.12-29-2011
20120101965TOPIC MODELS - Machine learning techniques may be used to train computing devices to understand a variety of documents (e.g., text files, web pages, articles, spreadsheets, etc.). Machine learning techniques may be used to address the issue that computing devices may lack the human intellect used to understand such documents, such as their semantic meaning. Accordingly, a topic model may be trained by sequentially processing documents and/or their features (e.g., document author, geographical location of author, creation date, social network information of author, and/or document metadata). Additionally, as provided herein, the topic model may be used to predict probabilities that words, features, documents, and/or document corpora, for example, are indicative of particular topics.04-26-2012
20100198763Reduction Of Classification Error Rates And Monitoring System Using An Artificial Class - Systems and methods for enhancing the accuracy of classifying a measurement by providing an artificial class. Seizure prediction systems may employ a classification system including an artificial class and a user interface for signaling uncertainty in classification when a measurement is classified in the artificial class.08-05-2010
20100198762AUTOMATED PREDICTIVE MODELING OF BUSINESS FUTURE EVENTS BASED ON HISTORICAL DATA - Predictive models are developed automatically for a plurality of modeling variables. The plurality of modeling variables is transformed, based on a transformation rule. A clustering of the transformed modeling variables is performed to create variable clusters. A set of variables is selected from the variable clusters based on a selection rule. A regression of the set of variables is performed to determine prediction variables. The prediction variables are utilized in developing a predictive model. The development of the predictive model may include modification of the predictive model, review of the plurality of transformations, and validation of the predictive model.08-05-2010
20100198760APPARATUS AND METHODS FOR MUSIC SIGNAL ANALYSIS - An apparatus for modelling layers in a music signal comprises a rhythm modelling module configured to model rhythm features of the music signal; a harmony modelling module configured to model harmony features of the music signal; and a music region modelling module configured to model music region features from the music signal.08-05-2010
20100198759Portal Performance Optimization - A method for portal performance optimization comprises receiving a request for a portal page, the portal page comprising a plurality of portlets; determining a current system load; determining, based on the current system load, whether a performance rule is triggered; and in the event a performance rule is triggered, deactivating at least one of the plurality of portlets. A system for portal performance optimization comprises a portal server configured to receive a request for a portal page, the portal page comprising a plurality of portlets, the portal server comprising a performance management component, the performance management component configured to determine a current system load; and a rules engine, the rules engine configured to determine if a performance rule is triggered by the determined current system load, and, in the event a performance rule is triggered, to apply the triggered performance rule to at least one of the plurality of portlets.08-05-2010
20100198757PERFORMANCE OF A SOCIAL NETWORK - Providing for characterizing and determining effectiveness of social networks is described herein. By way of example, data descriptive of inter-relationships of persons can be employed to generate a social connectivity map for users of a communication network. Data disseminated or consumed via the communication network can be monitored and characterized in conjunction with task performance. The characterization can be compared with a performance benchmark to rate a composition of a social network, or underlying network applications and functions, in effecting user tasks or other user activities. Accordingly, individuals and organizations can determine and compare the effectiveness of a network in assisting user activities based on predetermined benchmarks, which can be tuned to various aspects, functions or applications of an underlying social network.08-05-2010
20100198756METHODS AND SYSTEMS FOR MATCHING RECORDS AND NORMALIZING NAMES - Methods and systems are provided for normalizing strings and for matching records. In one implementation, a string is tokenized into components. Sequences of tags are generated by assigning tags to the components. A sequence of states is determined based on the sequences of tags. A normalized string is generated by normalizing the sequence of the states. A key record including key fields is extracted from a first data source. A candidate record including candidate fields is extracted from a second data source. A numerical record including numerical fields is computed by comparing the key fields and the candidate fields using comparison functions. Matching functions determined by an additive logistic regression method are applied to the numerical fields. Whether the key record and the candidate record are a match is determined based on a sum of results of the matching functions.08-05-2010
20130013537CLASSIFICATION METHOD AND APPARATUS - A method and system for building a classification model for classifying documents comprising: representing each of a plurality of documents by a vector of n dimensions, said n dimensions forming a vector space; and representing the classification of already classified documents into classes by separating said vector space into a plurality of subspaces by one or more hyperplanes.01-10-2013
20130013540GRAPH-BASED TRANSFER LEARNING - Transfer learning is the task of leveraging the information from labeled examples in some domains to predict the labels for examples in another domain. It finds abundant practical applications, such as sentiment prediction, image classification and network intrusion detection. A graph-based transfer learning framework propagates label information from a source domain to a target domain via the example-feature-example tripartite graph, and puts more emphasis on the labeled examples from the target domain via the example-example bipartite graph. An iterative algorithm renders the framework scalable to large-scale applications. The framework propagates the label information to both features irrelevant to the source domain and unlabeled examples in the target domain via common features in a principled way.01-10-2013
20130013539SYSTEM AND METHOD FOR DOMAIN ADAPTION WITH PARTIAL OBSERVATION - System, method and computer program product provides a novel domain adaption/transfer learning approach applied to the problem of classifying abbreviated documents, e.g., short text messages, instant messages, tweets. The proposed method uses a large number of multi-labeled examples (source domain) to improve the learning on the partial observations (target domain). Specifically, a hidden, higher-level abstraction space is learned that is meaningful for the multi-labeled examples in the source domain. This is done by simultaneously minimizing the document reconstruction error and the error in a classification model learned in the hidden space using known labels from the source domain. The partial observations in the target space are then mapped to the same hidden space, and classified into the label space determined by the source domain. Exemplary results provided for a Twitter dataset demonstrate that the method identifies meaningful hidden topics and provides useful classifications of specific tweets.01-10-2013
20130013538RECOVERING THE STRUCTURE OF SPARSE MARKOV NETWORKS FROM HIGH-DIMENSIONAL DATA - A method, information processing system, and computer readable article of manufacture model data. A first dataset is received that includes a first set of physical world data. At least one data model associated with the first dataset is generated based on the receiving. A second dataset is received that includes a second set of physical world data. The second dataset is compared to the at least one data model. A probability that the second dataset is modeled by the at least one data model is determined. A determination is made that the probability is above a given threshold. A decision associated with the second dataset based on the at least one data model is generated in response to the probability being above the given threshold. The probability and the decision are stored in memory. The probability and the decision are provided to user via a user interface.01-10-2013
20130013536METRIC LEARNING DEVICE, METRIC LEARNING METHOD, AND RECORDING MEDIUM - A metric learning device (01-10-2013
20130013535Method for Summarizing Event-Related Texts To Answer Search Queries - A method and apparatus for receiving training data that comprise a plurality of event-and-time-specific texts that are contextually related to a plurality of events; iteratively processing the training data to generate a modified network model that defines a plurality of states; receiving additional data that comprise a plurality of additional event-and-time-specific texts that are contextually related to a particular event; processing the additional data by applying the modified network model to the additional data to identify, within the plurality of additional event-and-time specific texts, a particular set of texts that belong to a particular state of the plurality of states; identifying, within the particular set of texts, one or more texts that are most representative of all texts in the particular set of texts that belong to the particular state; wherein the method is performed by one or more special-purpose computing devices.01-10-2013
20130013542SCALABLE TRAFFIC CLASSIFIER AND CLASSIFIER TRAINING SYSTEM - A traffic classifier has a plurality of binary classifiers, each associated with one of a plurality of calibrators. Each calibrator trained to translate an output score of the associated binary classifier into an estimated class probability value using a fitted logistic curve, each estimated class probability value indicating a probability that the packet flow on which the output score is based belongs to the traffic class associated with the binary classifier associated with the calibrator. The classifier training system configured to generate a training data based on network information gained using flow and packet sampling methods. In some embodiments, the classifier training system configured to generate reduced training data sets, one for each traffic class, reducing the training data related to traffic not associated with the traffic class.01-10-2013
20130013541System And Method For Invitation Targeting In A Web-Based Social Network - A system and method for selecting users of a web-based social network who are likely to respond to an invitation, each of the users having associated profile information is disclosed. The method includes selecting pilot users and a reduced set of keywords from the profile information. The method further includes sending the invitation to the pilot users, receiving responses from the pilot users, and classifying the responses as either positive or negative. A training set of vector pairs is created each vector pair representing a pilot user and including data representing a classified response and training keywords selected from the reduced set of keywords and associated profile information for the pilot user. A function is determined based on the vectors and used to calculate a likelihood that each of one or more users of the web based social network will respond to the invitation.01-10-2013
20120150774CONTROL APPARATUS - A control apparatus includes a learning portion which learns a control parameter by correcting a learning vector consisting of a plurality of variables and a control parameter based on a measurement vector. The control apparatus further includes an interpolation portion which computes the control parameter corresponding to current variables which represent a current environmental condition by interpolating the control parameter learned by the learning portion. The interpolation portion includes a selecting portion which selects three learning vectors from a plurality of learning vectors, and which computes the control parameter corresponding to the current variables by interpolating the control parameters on a flat surface including the selected three learning vectors.06-14-2012
20130018826LOCATION DETERMINATION USING GENERALIZED FINGERPRINTINGAANM Sundararajan; ArjunAACI RedmondAAST WAAACO USAAGP Sundararajan; Arjun Redmond WA USAANM Lin; Jyh-HanAACI Mercer IslandAAST WAAACO USAAGP Lin; Jyh-Han Mercer Island WA US - An RF fingerprinting methodology is generalized to include non-RF related factors. For each fingerprinted tile, there is an associated distance function between two fingerprints (the training fingerprint and the test fingerprint) from within that tile which may be a linear or non-linear combination of the deltas between multiple factors of the two fingerprints. The distance function for each tile is derived from a training dataset corresponding to that specific tile, and optimized to minimize the total difference between real distances and predicted distances. Upon receipt of an inference request, a result is derived from a combination of the fingerprints from the training dataset having the least distance per application of the distance function. Likely error for the tile is also determined to ascertain whether to rely on other location methods.01-17-2013
20130018824SENTIMENT CLASSIFIERS BASED ON FEATURE EXTRACTIONAANM Ghani; RayidAACI ChicagoAAST ILAACO USAAGP Ghani; Rayid Chicago IL USAANM Krema; MarkoAACI EvanstonAAST ILAACO USAAGP Krema; Marko Evanston IL US - Method and apparatus are provided for providing one or more sentiment classifiers from training data using supervised classification techniques based on features extracted from the training data. Training data includes a plurality of units such as, but not limited to, documents, paragraphs, sentences, and clauses. A feature extraction component extracts a plurality of features from the training data, and a feature value determination component determines a value for each extracted feature based on a frequency at which each feature occurs in the training data. On the other hand, a class labeling component labels each unit of the training data according to a plurality of sentiment classes to provide labeled training data. Thereafter, a sentiment classifier generation component provides a least one sentiment classifier based on the value of each extracted feature and the labeled training data using a supervised classification technique.01-17-2013
20130018823Detecting undesirable content on a social networkAANM Masood; Syed GhouseAACI Kuala LumpurAACO MYAAGP Masood; Syed Ghouse Kuala Lumpur MY - A method of detecting undesirable content on a social networking website. The method includes retrieving or accessing a post from a user's social networking page, identifying the content of a pre-defined set of features of the post, comparing the identified feature content with a database of known undesirable post feature content, and using the results of the comparison to determine whether the post is undesirable.01-17-2013
20130018825DETERMINATION OF A BASIS FOR A NEW DOMAIN MODEL BASED ON A PLURALITY OF LEARNED MODELSAANM GHANI; RayidAACI ChicagoAAST ILAACO USAAGP GHANI; Rayid Chicago IL USAANM Krema; MarkoAACI EvanstonAAST ILAACO USAAGP Krema; Marko Evanston IL US - In a machine learning system in which a plurality of learned models, each corresponding to a unique domain, already exist, new domain input for training a new domain model may be provided. Statistical characteristics of features in the new domain input are first determined. The resulting new domain statistical characteristics are then compared with statistical characteristics of features in prior input previously provided for training at least some of the plurality of learned models. Thereafter, at least one learned model of the plurality of learned models is identified as the basis for the new domain model when the new domain input statistical characteristics compare favorably with the statistical characteristics of the features in the prior input corresponding to the at least one learned model.01-17-2013
20130018828SYSTEM AND METHOD FOR AUTOMATED LABELING OF TEXT DOCUMENTS USING ONTOLOGIES - A first mapping function automatically maps a plurality of documents each with a concept of ontology to create a documents-to-ontology distribution. An ontology-to-class distribution that maps concepts in the ontology to class labels, respectively, is received, and a classifier is generated that labels a selected document with an associated class identified based on the documents-to-ontology distribution and the ontology-to-class distribution.01-17-2013
20130018827SYSTEM AND METHOD FOR AUTOMATED LABELING OF TEXT DOCUMENTS USING ONTOLOGIESAANM He; JingruiAACI OssiningAAST NYAACO USAAGP He; Jingrui Ossining NY USAANM Lawrence; Richard D.AACI RidgefieldAAST CTAACO USAAGP Lawrence; Richard D. Ridgefield CT USAANM Melville; PremAACI White PlainsAAST NYAACO USAAGP Melville; Prem White Plains NY USAANM Sindhwani; VikasAACI HawthorneAAST NYAACO USAAGP Sindhwani; Vikas Hawthorne NY USAANM Chenthamarakshan; Vijil E.AACI OssiningAAST NYAACO USAAGP Chenthamarakshan; Vijil E. Ossining NY US - A first mapping function automatically maps a plurality of documents each with a concept of ontology to create a documents-to-ontology distribution. An ontology-to-class distribution that maps concepts in the ontology to class labels, respectively, is received, and a classifier is generated that labels a selected document with an associated class identified based on the documents-to-ontology distribution and the ontology-to-class distribution.01-17-2013
20110161262PROFILE CONFIGURATION FOR A MOBILE COMPUTING DEVICE - Data processing apparatus is disclosed comprising: a sensor module configured to sense a first profile comprised of one or more attributes of an environment of said data processing apparatus; and a classification module configured to assign a prediction factor to each of said one or more attributes of said first profile and to store each said attribute and assigned prediction factor as a stored profile.06-30-2011
20110161261METHOD AND SYSTEM FOR TRAFFIC PREDICTION BASED ON SPACE-TIME RELATION - A system and method for traffic prediction based on space-time relation are disclosed. The system comprises a section spatial influence determining section for determining, for each of a plurality of sections to be predicted, spatial influences on the section by its neighboring sections; a traffic prediction model establishment section for establishing, for each of the plurality of sections to be predicted, a traffic prediction model by using the determined spatial influences and historical traffic data of the plurality of sections; and a traffic prediction section for predicting traffic of each of the plurality of sections to be predicted for a future time period by using real-time traffic data and the traffic prediction model. An apparatus and method for determining spatial influences among sections, as well as an apparatus and method for traffic prediction, are also disclosed. With the present invention, a spatial influence of a section can be used as a spatial operator and a time sequence model can be incorporated, such that the influences on a current section by its neighboring section for a plurality of spatial orders can be taken into account. In this way, the traffic condition in a spatial scope can be measured more practically, so as to improve accuracy of prediction.06-30-2011
20110161260USER-DRIVEN INDEX SELECTION - Techniques for index building are described. Clickcounts of respective training URLs may indicate a number of times that corresponding training URLs were clicked in search engine results. A machine learning algorithm implemented on a computer computes a trained model that is then stored. The clickcounts and respective URLs are passed to the machine learning algorithm to train the model to predict probabilities based on feature vectors of URLs. An index of web pages is built for a set of URLs that identify the web pages. Feature vectors for the URLs are computed. Probabilities of the web pages of the URLs being searched in the future by users may be computed by processing the feature vectors with the trained model. The probabilities may be used to determine which of the URLs to include in the index.06-30-2011
20110161259SYSTEM AND METHOD FOR SIMPLIFICATION OF A MATRIX BASED BOOSTING ALGORITHM - A method for simplification of a matrix based boosting algorithm divides a feature set comprising a plurality of feature data into several subsets, and assigns a number to each subset. The method selects a plurality of number groups including N subsets randomly. The method further computes a value by boosting algorithm according to each of the number groups for obtaining an acceptable false positive value.06-30-2011
20130024404FLIGHT CACHING METHODS AND APPARATUS - According to some aspects, a system is provided comprising at least one computer readable storage medium storing a cache of flight information comprising a plurality of flight solutions, the cache capable of being accessed to obtain flight solutions that meet a criteria specified in one or more flight search queries, and at least one computer programmed to apply at least one machine learning model to at least some of the flight information in the flight information cache to classify at least one of the plurality of flight solutions according to an assessed fidelity of the at least one flight solution, and perform at least one action based on the classified at least one flight solution.01-24-2013
20130024406Scalable Ontology Extraction - Techniques for facilitating learning of one or more ontological rules of a resource description framework database are provided. The techniques include obtaining ontology vocabulary from a resource description framework database, generating a rule hypothesis by incrementally building upon a previously learnt rule from the database by adding one or more predicates to the previously learnt rule, performing a constraint check on the generated rule hypothesis by determining compatibility with each previously learnt rule to ensure that a complete rule set including each previously learnt rule and the generated rule hypothesis is consistent, validating the rule hypothesis as a rule using one or more association rule mining techniques to determine validity of the rule hypothesis against the database, and applying the rule to the database to infer one or more facts from the database to facilitate learning of one or more additional ontological rules.01-24-2013
20130024405DISTRIBUTED SCALABLE INCREMENTALLY UPDATED MODELS IN DECISIONING SYSTEMS - In one embodiment, first weight information indicating a first set of delta values is obtained, where the first set of delta values includes a first delta value for each weight in a set of weights, the set of weights including a weight for each of a set of one or more parameters of a model. In addition, second weight information indicating a second set of delta values is obtained, where the second set of delta values includes a second delta value for each weight in the set of weights. Combined weight information including a combined set of delta values or a combined set of weights is generated based, at least in part, upon the first weight information and the second weight information.01-24-2013
20130024403AUTOMATICALLY INDUCED CLASS BASED SHRINKAGE FEATURES FOR TEXT CLASSIFICATION - A method and apparatus are provided for automatically inducing class based shrinkage features. The method includes clustering each word in a set of word groupings of a given type into a respective one of a plurality of classes. The method further includes selecting and extracting a set of class-based shrinkage features from the set of word groupings based on the plurality of classes. The set of class-based shrinkage features is specifically selected for an intended classification application.01-24-2013
20130173505System and Method For Artificial Lift System Analysis - A system, method, and computer program product are disclosed for analysis of artificial lift systems. An Artificial Lift Analysis Solution (ALAS) is also provided to view and analyze artificial lift well data trends, prediction and detection event alerts, and to diagnose system conditions to facilitate production optimization. Production well information is provided for a plurality of the production wells each being associated with an artificial lift system. Artificial lift system failure alerts for the plurality of production wells are received and processed on a computer. A relevance measure for each of the artificial lift system failure alerts is determined responsive to the production well information. A summary of the artificial lift system failure alerts is displayed in an ordering based on the relevance measure.07-04-2013
20130173506HYBRID LOCATION USING PATTERN RECOGNITION OF LOCATION READINGS AND SIGNAL STRENGTHS OF WIRELESS ACCESS POINTS - A query device scans radio frequencies for visible transmitting devices. The querying device receives at least a signal strength and identifier information associated with each of the transmitting devices. The list of visible devices is used to query a database containing location information for a plurality of visible devices. The list may be sent to a locationing system that may perform a location analysis on the resulting data to return a location to the query device. The weighted average of the locations returned in the database query may be computed to determine the location of the querying device, with the weight for each of the locations being the current signal strength detected by the querying device. Neural network analysis may also be used to determine the location of the querying device. Learning and seeding operations many also be used to populate the database with location information for transmitting devices.07-04-2013
20130173507ADAPTIVE CUSTOMIZED PRESENTATION OF BUSINESS INTELLIGENCE INFORMATION - In one example, a method includes receiving information on a user role of a user account associated with a business intelligence system. The method further includes gathering information on interactions of the user account with the business intelligence system. The method further includes generating an initial business intelligence output based on data from one or more data sources. The method further includes generating a customized business intelligence output for the user account based on the initial business intelligence output, the user role, and the interactions of the user account with the business intelligence system. The method further includes providing the customized business intelligence output to the user account.07-04-2013
20080243731GENERALIZED SEQUENTIAL MINIMAL OPTIMIZATION FOR SVM+ COMPUTATIONS - A system and method for support vector machine plus (SVM+) computations include selecting a set of indexes for a target function to create a quadratic function depending on a number of variables, and reducing the number of variables to two in the quadratic function using linear constraints. An extreme point is computed for the quadratic function in closed form. A two-dimensional set is defined where the indexes determine whether a data point is in the two-dimensional set or not. A determination is made of whether the extreme point belongs to the two-dimensional set. If the extreme point belongs to the two-dimensional set, the extreme point defines a maximum and defines a new set of parameters for a next iteration. Otherwise, the quadratic function is restricted on at least one boundary of the two-dimensional set to create a one-dimensional quadratic function. The steps are repeated until the maximum is determined.10-02-2008
20080235166TRAINING A MODEL OF A NON-LINEAR PROCESS - System and method for modeling a nonlinear process. A combined model for predictive optimization or control of a nonlinear process includes a nonlinear approximator, coupled to a parameterized dynamic or static model, operable to model the nonlinear process. The nonlinear approximator receives process inputs, and generates parameters for the parameterized dynamic model. The parameterized dynamic model receives the parameters and process inputs, and generates predicted process outputs based on the parameters and process inputs, where the predicted process outputs are useable to analyze and/or control the nonlinear process. The combined model may be trained in an integrated manner, e.g., substantially concurrently, by identifying process inputs and outputs (I/O), collecting data for process I/O, determining constraints on model behavior from prior knowledge, formulating an optimization problem, executing an optimization algorithm to determine model parameters subject to the determined constraints, and verifying the compliance of the model with the constraints.09-25-2008
20080235164Apparatus, method and computer program product providing a hierarchical approach to command-control tasks using a brain-computer interface - Disclosed is a method, a computer program product, and a device that are responsive to detected mental states of a user to perform selection processes to execute a task. The method includes providing a hierarchical multi-level decision tree structure comprised of internal nodes and leaf nodes, where the decision tree structure represents a task. The method further includes navigating, using information derived from detected mental states of the user, through levels of the decision tree structure to reach a leaf node to accomplish the task. The step of navigating includes selecting, using the information derived from the detected mental states of the user, between attribute values associated with internal nodes of the decision tree structure. As non-limiting examples, the device may be a communication device, and the task may be a name dialing or a command/control task.09-25-2008
20130173509METHOD AND ARRANGEMENT FOR PROCESSING DATA - A method and arrangement for processing data when training a data model involving multiple iterations of data records in a dataset (07-04-2013
20110246401GRAPHICAL INFORMATION NAVIGATOR - Embodiments are disclosed for facilitating graphical navigation of data. In a specific embodiment, the system includes a graphical user interface that is adapted to graphically depict data via one or more displayed icons. The graphical user interface is further adapted to enable a user to cause display a first icon and one or more second icons associated therewith by selection of the first icon. A learning module is adapted to monitor use of the graphical user interface and to adjust behavior of the graphical user interface in response thereto based on learned information obtained from monitoring the use of the graphical user interface. The system may be specifically adapted to facilitate user navigation of data that is maintained by Enterprise Resource Planning (ERP) software.10-06-2011
20110246400SYSTEM FOR OPTICAL METROLOGY OPTIMIZATION USING RAY TRACING - Provided is a system for determining profile parameters of a sample structure on a workpiece using an optical metrology system optimized to achieve one or more accuracy targets. The optical metrology system comprises an optical metrology tool configured to measure a diffraction signal off a sample structure, an optical metrology tool model configured to model the optical metrology tool using a selected number of rays and selected beam propagation parameters for the illumination beam and the diffraction beam; a signal adjuster configured to adjust the measured diffraction signal off the sample structure using the optical metrology tool model and calibration parameters, the signal adjuster generating an adjusted metrology output signal; and a profile extractor configured to determine one or more profile parameters of the sample structure using the adjusted metrology output signal, a profile model of the sample structure, and one or more extraction modules.10-06-2011
20130144820METHOD OF LEARNING A CONTEXT OF A SEGMENT OF TEXT, AND ASSOCIATED HANDHELD ELECTRONIC DEVICE - An improved method of learning a context of a segment of text input enables facilitated text input on an improved handheld electronic device. In response to a series of inputs, segments and other objects are analyzed to generate a proposed character interpretation of the series of inputs. Responsive to detecting a replacement of a segment of the character interpretation with another segment, a combination object comprising the another segment and a preceding object is stored. In response to another series of inputs, the combination object can be employed by a processing algorithm to ascertain a preference for the another segment in the context of the preceding object of the combination object.06-06-2013
20130144819SCORE NORMALIZATION - Methods, systems, and apparatus, including computer programs encoded on computer storage media, for score normalization. One of the methods includes receiving initial training data, the initial training data comprising initial training records, each initial training record identifying input data as input and a category as output. The method includes generating a first trained predictive model using the initial training data and a training function. The method includes generating intermediate training records by inputting input data of the initial training records to a second trained predictive model, the second trained predictive model generated using the training function, each intermediate training record having a score. The method also includes generating a score normalization model using a score normalization training function and the intermediate training records.06-06-2013
20130144818NETWORK INFORMATION METHODS DEVICES AND SYSTEMS - Methods and systems for predicting links in a network, such as a social network, are disclosed. The existing network structure can be used to optimize link prediction. The methods and systems can learn a distance metric and/or a degree preference function that are structure preserving to predict links for new/existing nodes based on node properties.06-06-2013
20130144817Parallel training of a Support Vector Machine (SVM) with distributed block minimization - A method to solve large scale linear SVM that is efficient in terms of computation, data storage and communication requirements. The approach works efficiently over very large datasets, and it does not require any master node to keep any examples in its memory. The algorithm assumes that the dataset is partitioned over several nodes on a cluster, and it performs “distributed block minimization” to achieve the desired results. Using the described approach, the communication complexity of the algorithm is independent of the number of training examples.06-06-2013
20130144816HEALTH CARE INCIDENT PREDICTION - Embodiments described herein relate to apparatuses and methods for incident prediction alerts for transmission to a health care organization system by applying rules to data sets. Each rule may define a set of data elements linked to an incident, and a processor may detect one or more sets of data elements in the data sets. The processor may normalize the data feeds, generate rules on historical data, generate prediction alerts by applying rules to near-real time data feeds, train to update rules, validate and error check rules, remove statistical noise, generate visualizations for the data feeds, and receive input and feedback data.06-06-2013
20130144815MAKING PREDICTIONS REGARDING EVALUATION OF FUNCTIONS FOR A DATABASE ENVIRONMENT - A prediction regarding one or more functions can be made for a database environment. In particular, a predication can be made with respect to values stored in at least one column of at least one table in a database, based on the evaluation of one or more functions for a subset of possible column values (i.e., resultant values derived from the evaluation of a subset of possible column values) without the need to calculate the function(s) for all of the actual entries in the column of the table(s). In effect, a functional predicate can be transformed (or translated) to a predicate that is dependent on the column values instead of the evaluation of one or more functions for the column values.06-06-2013
20130144814CONDITIONAL PROBABILITY OPERATOR FOR EVENT PROCESSING SYSTEMS - An event processing system in which a computer receives an input event comprising one or more factors. The computer evaluates the factors of the input event based on an event processing rule containing a pattern detection operator and a conditional probability operator. The conditional probability operator can operate to calculate a conditional probability for a set of training data that a specified pattern will appear in the factors of an input event given a specified output event, and can further operate to assign a conditional rule value a binary value based on how the conditional probability compares to a target probability.06-06-2013
20130144813Analyzing Data Sets with the Help of Inexpert Humans to Find Patterns - A combined computer/human approach is used to detect actionable insights in large data sets. Automated computer analysis used to identify patterns (e.g., possibly meaningful patterns or subsets within the data). These are presented to humans for feedback, where the humans may have little to no training in the statistical methods used to detect actionable insights. Feedback from the humans is used to improve the pattern detection and facilitate the detection of actionable insights.06-06-2013
20130144812PROBABILISTIC MODEL APPROXIMATION FOR STATISTICAL RELATIONAL LEARNING - Various technologies described herein pertain to approximating an inputted probabilistic model for statistical relational learning. An initial approximation of formulae included in an inputted probabilistic model can be formed, where the initial approximation of the formulae omits axioms included in the inputted probabilistic model. Further, an approximated probabilistic model of the inputted probabilistic model can be constructed, where the approximated probabilistic model includes the initial approximation of the formulae. Moreover, the approximated probabilistic model and evidence can be fed to a relational learning engine, and a most probable explanation (MPE) world can be received from the relational learning engine. The evidence can comprise existing valuations of a subset of relations included in the inputted probabilistic model. The MPE world can include valuations for the relations included in the inputted probabilistic model. The MPE world can be outputted when the input probabilistic model lacks an axiom violated by the MPE world.06-06-2013
20130173503COMPOUND SELECTION IN DRUG DISCOVERY - Methods and systems for determining the selection criteria that in its embodiments can distinguish compounds that successfully meet an objective from those that do not, determine the importance of selection criterion in selecting test compounds that have a high probability of achieving an objective and automatically apply the selection criteria to select test compounds with a high chance of meeting an objective.07-04-2013
20130173508DEFECT CLASSIFICATION APPARATUS - The present invention has its objective to provide a defect classification apparatus which suppresses over-fitting and accurately classify the defect type of a defect. A defect classification apparatus is provided in which a data point indicating feature information of a defect to be classified having an unknown defect type is mapped to a point in a mapping space having a dimensional number higher than the number of features constituting the feature information, and the defect type of the defect to be classified is classified based on in which of two regions of defect type, which are formed by separating the mapping space by a decision boundary, the mapped point is located, wherein a discriminant function indicating the decision boundary is determined by adopting a weight which minimizes the sum of the classification error, which corresponds to the accuracy in classifying a training defect dataset, and a regularization term, which has a positive correlation with the dimensional number of the decision boundary, as the weight for each feature constituting the discriminant function.07-04-2013
20130173504SYSTEMS AND METHODS FOR ACTION RECOGNITION - Systems provided herein include a learning environment and an agent. The learning environment includes an avatar and an object. A state signal corresponding to a state of the learning environment includes a location and orientation of the avatar and the object. The agent is adapted to receive the state signal, to issue an action capable of generating at least one change in the state of the learning environment, to produce a set of observations relevant to a task, to hypothesize a set of action models configured to explain the observations, and to vet the set of action models to identify a learned model for the task.07-04-2013
20120254081OPTIMIZATION CONTROL SYSTEM - Provided is an optimization control system in an attempt to improve searching accuracy of an optimal solution defining a behavior mode for a control subject. A plan storing element 10-04-2012
20120254080Markov Modeling of Service Usage Patterns - A system for analyzing service usage utilizing Markov models. Records of client requests to the service are extracted from at least one log. The records are grouped by client and sorted by timestamp. A pattern of requests that form an action is detected. Each action has a time. A probability is calculated of a transition from a precedent action to a subsequent action, where the precedent action has a time prior to the subsequent action. A delay time is also calculated between a precedent action and a subsequent action. A probability is calculated for a delay time, such as the likelihood that a delay from a precedent action to a subsequent action will fall within a given time interval.10-04-2012
20120254076SUPERVISED RE-RANKING FOR VISUAL SEARCH - Supervised re-ranking for visual search may include re-ordering images that are returned in response to a text-based image search by exploiting visual information included in the images. In one example, supervised re-ranking for visual search may include receiving a textual query, obtaining an initial ranking result including a plurality of images corresponding to the textual query, and representing the textual query by a visual context of the plurality of images. A query-independent re-ranking model may be trained based on visual re-ranking features of the plurality of images of the textual query in accordance with a supervised training algorithm.10-04-2012
20130091081LATENT FACTOR DEENDENCY STRUCTURE DETERMINATION - Disclosed is a general learning framework for computer implementation that induces sparsity on the undirected graphical model imposed on the vector of latent factors. A latent factor model SLFA is disclosed as a matrix factorization problem with a special regularization term that encourages collaborative reconstruction. Advantageously, the model may simultaneously learn the lower-dimensional representation for data and model the pairwise relationships between latent factors explicitly. An on-line learning algorithm is disclosed to make the model amenable to large-scale learning problems. Experimental results on two synthetic data and two real-world data sets demonstrate that pairwise relationships and latent factors learned by the model provide a more structured way of exploring high-dimensional data, and the learned representations achieve the state-of-the-art classification performance04-11-2013
20130091080PROBABILISTIC MODEL CHECKING OF SYSTEMS WITH RANGED PROBABILITIES - Systems and methods for model checking of live systems are shown that include learning an interval discrete-time Markov chain (IDTMC) model of a deployed system from system logs; and checking the IDTMC model with a processor to determine a probability of violating one or more probabilistic safety properties. Checking the IDTMC model includes calculating a linear part exactly using affine arithmetic; and over-approximating a non-linear part using interval arithmetic.04-11-2013
20130091079USING A HEURISTICALLY-GENERATED POLICY TO DYNAMICALLY SELECT STRING ANALYSIS ALGORITHMS FOR CLIENT QUERIES - A method for dynamically selecting string analysis algorithms can begin with the training of the dynamic string analysis handler of a string analysis module to effectively handle a subset of string queries having contextual metadata received from a client application in an instructional environment. The effectiveness of the training module can be based upon feedback from the client application. Upon completion of the training, a string analysis algorithm selection policy can be synthesized. The string analysis algorithm selection policy can correlate a context of a string query in the subset to the usage of a string analysis algorithm. When in the operational environment, the dynamic string analysis handler can dynamically handle string queries having contextual metadata received from the client application in accordance with the string analysis algorithm selection policy. The string analysis algorithm to be used for a string query can be dynamically and independently determined.04-11-2013
20130091078Method And Apparatus To Determine Rules Implementation Decision - A technique and associated mechanism that guides the user through a set of questions relating to operation rules used in the design of Service Oriented Architecture Systems (SOAs). The questions are related to key aspects of a solution—security, maintenance frequency, usage demand/performance and complexity. Preferably, the questions are yes-or-no questions. Based on the answers provided, an appropriate path will be selected categorize into an appropriate category. The category of the rule will require, or at least suggest, the SOA component into which the rule will be implemented when it is implemented by the SOA designer. the technique is technology specific agnostic and helps in selecting an appropriate tool/platform in a standard and consistent manner.04-11-2013
20130179375Signal Detection Algorithms to Identify Drug Effects and Drug Interactions - An algorithm according to an embodiment of the present invention provides for latent signal detection of adverse events. Embodiments infer the presence of adverse drug events from large observational databases housed by the FDA, WHO, and other governmental organizations. The disclosed algorithms do not require the adverse event to be reported explicitly. Instead, the algorithms infer the presence of adverse events through more common secondary effects. In an embodiment, machine learning techniques are used for this purpose.07-11-2013
20130097105CONTEXT AWARE APPARATUS AND METHOD - A context aware apparatus is provided. The context aware apparatus includes an extracting unit configured to extract a terminological-box (T-box) from a semantic model, a first generating unit configured to generate a reasoning rule based on the extracted T-box, a second generating unit configured to generate a first assertion-box (A-box) based on sensing information, and a reasoning unit configured to infer a user context based on the reasoning rule and the first A-box.04-18-2013
20130097107INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD, AND PROGRAM - Provided is an information processing apparatus including: a reward estimator generating unit using action history data, which includes state data expressing a state of an agent, action data expressing an agent's action, and a reward value expressing a reward of the action, as learning data to generate, through machine learning, a reward estimator estimating the reward value from inputted state data and action data; an action selecting unit preferentially selecting an action not included in the action history data but with a high estimated reward value; and an action history adding unit causing the agent to perform the selected action and adding to the action history data the state data and action data for the action and the action's reward value in association with each other. The reward estimator is regenerated when a set of state data, action data, and the reward value is added to the action history data.04-18-2013
20130097106APPARATUS AND METHOD FOR EXTENDING A DEFAULT MODEL OF A TERMINAL - Provided are an apparatus and method for extending a default model of a terminal. The apparatus may extend a default model of the terminal using an extension model relating to the default model and using linked data.04-18-2013
20130097104METHOD AND SYSTEM FOR DOCUMENT CLASSIFICATION - The present invention provides a method for document classification, especially an adaptive learning method for document classification. The document includes a plurality of feature words. The method includes steps of calculating a plurality of similarities between the document and a categorical basic knowledge; calculating a first ratio of a first largest similarity to a second largest similarity of the plurality of similarities; storing the feature words of the document as an extensive categorical knowledge when the first ratio is larger than a first threshold value; and updating the categorical basic knowledge by using the extensive categorical knowledge.04-18-2013
20130097102SYSTEMS AND METHODS FOR MANAGING PUBLICATION OF ONLINE ADVERTISEMENTS - Exemplary embodiments provide systems, devices, one or more non-transitory computer-readable media and computer-executable methods for managing publication of online advertising. In exemplary embodiments, computer-based publication techniques may include, but is not limited to, automatically determining whether the content of a particular web page article is suitable or unsuitable for accompaniment with one or more advertisements, automatically determining whether an advertisement is suitable or unsuitable for publication on a web page associated with a web page article, and automatically determining a category that may be used to classify the content of a web page article in order to select one or more categories of advertisements suitable for accompaniment with the web page article.04-18-2013
20130097103Techniques for Generating Balanced and Class-Independent Training Data From Unlabeled Data Set - Techniques for creating training sets for predictive modeling are provided. In one aspect, a method for generating training data from an unlabeled data set is provided which includes the following steps. A small initial set of data is selected from the unlabeled data set. Labels are acquired for the initial set of data selected from the unlabeled data set resulting in labeled data. The data in the unlabeled data set is clustered using a semi-supervised clustering process along with the labeled data to produce data clusters. Data samples are chosen from each of the clusters to use as the training data. The selecting, presenting, clustering and choosing steps are repeated with one or more additional sets of data selected from the unlabeled data set until a desired amount of training data has been obtained, wherein at each iteration an amount of the labeled data is increased.04-18-2013
20130103619COMPOSITE PRODUCTION RULES - A method for forming and using a composite production rule may include compiling, by a computer system, a decision table or a decision tree to generate a composite production rule. The method may also include generating the composite production rule and selecting, by the computer system, an algorithm for compiling the composite production rule. The method may additionally include compiling, by the computer system, the composite production rule into an executable program based on pattern matching of the selected algorithm. The method may further include executing, by the computer system, the composite production rule to provide an output based on the composite production rule.04-25-2013
20130103625INFORMATION PROCESSING APPARATUS AND METHOD, AND PROGRAM THEREOF - There is provided an information processing apparatus including: evaluation information extracting means extracting evaluation information from evaluation of every user for an item; preference information creating means for creating preference information indicating a preference of every user on the basis of the evaluation information extracted by the evaluation information extracting means and an item characteristic amount indicating a characteristic of the item; space creating means for creating a space in which the user is located, according to the preference information; and display control means for controlling display of the user located in the space, according to the space created by the space creating means and the preference information. The apparatus may be applied to, for example, an image display apparatus which displays server images for providing a variety of items and information.04-25-2013
20130103623Computer-Implemented Systems and Methods for Detection of Sentiment in Writing - Systems and methods are provided for the detection of sentiment in writing. A plurality of texts is received from a larger collection of writing samples with a computer system. A set of seed words from the plurality of texts are labeled as being of positive sentiment or of negative sentiment with the computer system. The set of seed words is expanded in size with the computer system to provide an expanded set of seed words. Intensity values are assigned to words of the expanded set of seed words. Each of the words of the expanded set of seed words is assigned three intensity values: a value corresponding to the strength of the word's association with a positive polarity class, a value corresponding to the strength of the word's association with a negative polarity class, and a value corresponding to the strength of the word's association with a neutral polarity class.04-25-2013
20130103622AUTOMATED CONTROL-SCHEDULE ACQUISITION WITHIN AN INTELLIGENT CONTROLLER - The current application is directed to intelligent controllers that initially aggressively learn, and then continue, in a steady-state mode, to monitor, learn, and modify one or more control schedules that specify a desired operational behavior of a device, machine, system, or organization controlled by the intelligent controller. An intelligent controller generally acquires one or more initial control schedules through schedule-creation and schedule-modification interfaces or by accessing a default control schedule stored locally or remotely in a memory or mass-storage device. The intelligent controller then proceeds to learn, over time, a desired operational behavior for the device, machine, system, or organization controlled by the intelligent controller based on immediate-control inputs, schedule-modification inputs, and previous and current control schedules, encoding the desired operational behavior in one or more control schedules and/or sub-schedules.04-25-2013
20130103621INTELLIGENT CONTROLLER PROVIDING TIME TO TARGET STATE - The current application is directed to intelligent controllers that continuously, periodically, or intermittently calculate and display the time remaining until a control task is projected to be completed by the intelligent controller. In general, the intelligent controller employs multiple different models for the time behavior of one or more parameters or characteristics within a region or volume affected by one or more devices, systems, or other entities controlled by the intelligent controller. The intelligent controller collects data, over time, from which the models are constructed and uses the models to predict the time remaining until one or more characteristics or parameters of the region or volume reaches one or more specified values as a result of intelligent controller control of one or more devices, systems, or other entities.04-25-2013
20130103620FEATURE VECTOR CLASSIFICATION DEVICE AND METHOD THEREOF - Disclosed is a feature vector classification device which includes an initial condition setting unit; a variable calculating unit configured to receive a training vector and to calculate an error and a weight according to setting of the initial condition setting unit; a loop deciding unit configured to determine whether re-calculation is required, based on a comparison result between the calculated error and an error threshold; and a hyperplane generating unit configured to generate a hyperplane when an end signal is received from the loop deciding unit.04-25-2013
20130103617Computer-Implemented Systems And Methods For Forecasting And Estimation Using Grid Regression - Systems and methods are provided for estimating a value for a target variable. A plurality of known entities are assigned to cells of a grid, where the known entities are assigned to the cells based upon attribute data. A determination is made as to whether each cell has at least a threshold number of assigned known entities. When one of the cells contains fewer than the threshold number of known entities, cells are combined to form a super cell. A model is generated for each cell and super cell based upon target variable values for known entities assigned to that cell or super cell. Data for a target entity is received, and the target entity is assigned to one the cells. One of the models is selected based upon the cell assignment, and an estimate is generated for the target variable for the target entity using the selected model.04-25-2013
20130103618DECISION MAKING WITH ANALYTICALLY COMBINED SPLIT CONDITIONS - Systems, methods, and other embodiments associated with decision making with analytically combined split conditions are provided. In one embodiment, a method for classifying data is provided. An input data sample is received for classification as belonging to one of two possible classes. The input data sample includes a set of attribute values. The method includes evaluating the set of attribute values with a tree function that defines a decision boundary of a classification tree. The tree function classifies an input data sample as belonging to one of the two possible classes based, at least in part, on the attribute values of the input data sample. In another embodiment parameters of the tree function are derived by applying a gradient descent parameter update rule to the training data samples.04-25-2013
20110313960GRAPH PATTERN RECOGNITION INTERFACE - In some example embodiments, a system and method are illustrated as including receive pattern data that includes transaction data relating to transactions between persons. Next, the system and method may include building at least one secondary network based upon the pattern data. Additionally, the system and method may include displaying the at least one secondary network.12-22-2011
20110313959METHOD AND SYSTEM FOR LISTING CATEGORIZATION - Embodiments of a method and system for listing categorization are disclosed. A category structure may be accessed. The category structure may include a plurality of categories for items. A set of training data may be accessed from a plurality of listings from at least one of supply data and/or demand data. The supply data may be generated from seller activity of a plurality of users in a networked system. The demand data may be generated from buyer activity of the plurality of users in the networked system. Each listing may include a category from the category structure. The set of training data may be provided to a categorization application for training. The categorization application may be capable of building listing statistics by applying a classifier to the set of training data and recommending a category from the category structure for a new listing by utilizing the listing statistics.12-22-2011
20110313958SYSTEM AND METHOD FOR EMPIRICAL ENSEMBLE-BASED VIRTUAL SENSING OF PARTICULATES - A virtual sensor system and method for the estimation of an amount or concentration of particulate matter resulting from natural or man made processes comprising two or more empirical models arranged for being trained using empirical data from the processes, for receiving one or more signal input values from one or more sensors of the processes and calculating a signal output value based on the signal input values where the signal output value represents an intermediate amount or concentration of particulate matter. Further a combination function is arranged for receiving the signal output values and continuously calculating the amount or concentration of PM.12-22-2011
20110313957DATA PROCESSING APPARATUS, DATA PROCESSING METHOD AND PROGRAM - A data processing apparatus includes: a learning section which obtains parameters of a probability model; a destination and stopover estimating section which estimates a destination node corresponding to a movement destination and a stopover node corresponding to a movement stopover; a current location estimating section which inputs the movement history data of the user within a predetermined time from a current time to the probability model using the parameters obtained by learning, and estimates a current location node corresponding to a current location of the user; a searching section which searches for a route to the destination from the current location of the user; and a calculating section which calculates an arrival probability and a time to reach the searched destination. The learning section includes a known or unknown determining section, a parameter updating section, a new model generating section, and a new model combining section.12-22-2011
20110313956INFORMATION PROCESSING APPARATUS, INFORMATION PROCESSING METHOD AND PROGRAM - There is provided an apparatus including an information processing apparatus, including a behavior learning unit that learns an activity model representing an activity state of a user as a probabilistic state transition model from time-series data of the user's location, and that finds a state node corresponding to a location where the user conducts activities using the user's activity model, a candidate assigning unit that assigns category candidates related to location or time to the state node, and a display unit that presents the category candidate to the user.12-22-2011
20110313955REAL-TIME INTELLIGENT VIRTUAL CHARACTERS WITH LEARNING CAPABILITIES - A system, method, and computer-readable instructions for real-time characters with learning capabilities. A plurality of rules are defined in a rules-based system, each of the rules defining a condition that determines a behavior of a virtual agent when the rule is triggered by the condition being satisfied so that upon triggering of multiple rules at the same time, each of the behaviors of the multiple rules whose conditions were satisfied are combined into a resultant behavior for the virtual agent. This resultant behavior is compared with a desired behavior to providing feedback in the form of rewards or punishments to each of the multiple rules based on their corresponding contribution to the resultant behavior as compared to the desired behavior.12-22-2011
20110313954COMMUNITY MODEL BASED POINT OF INTEREST LOCAL SEARCH - The present disclosure describes a community model based point of interest local search platform. Specifically, logs of users that store selections while accessing a point of interest application are loaded into a database. The logs are of users that have similar demographic or other community attributes. The logs are then mined for contextual parameters, including, but not limited to time of day, day of week, distance, activity, environment, popularity, and personal preferences. The point of interest selections are then mapped to a multi-dimensional map where each dimension corresponds to a contextual parameter. Clusters are evaluated by a classifier and classes of users of the community are identified. When a user then queries the community model based point of interest local search platform, contextual parameters are submitted with the query, relevant classes identified, and the corresponding point of interest information is displayed to the user.12-22-2011
20110313953Automated Classification Pipeline Tuning Under Mobile Device Resource Constraints - An architecture and techniques to enable a mobile device to efficiently classify raw sensor data into useful high level inferred data is discussed. Classification efficiency is achieved by tuning the mobile device's raw sensor data classification pipeline to attain a balance of accuracy, latency and energy suitable for mobile devices. The tuning of the classification pipeline is accomplished via a multi-pipeline tuning approach that uses Statistical Machine Learning Tools (SMLTs) and a classification cost modeler.12-22-2011
20130103624Method and system for estimating response to token instance of interest - Estimating a response to a token instance of interest, including the steps of: receiving token instances to which a user was exposed, receiving a total response of the user to the token instances; receiving attention levels of the user in the token instances; selecting the token instance of interest from among the token instances based on the attention level; and estimating the response to the token instance of interest from the total response.04-25-2013
20130124439INFORMATION EXTRACTION SYSTEM, METHOD, AND PROGRAM - An information extraction system includes: solution request sentence set acquisition means for acquiring a sentence set matching a positive example solution request pattern which represents a positive example of a sentence including a problem evoking expression and a sentence set matching a negative example solution request pattern representing an opposite request to the positive example solution request, from a corpus respectively as a positive example solution request sentence set and a negative example solution request sentence set, and associating parts, in the positive example solution request sentence set and the negative example solution request sentence set, that correspond to the problem evoking expression in the positive example solution request pattern and the negative example solution request pattern, with a positive example and a negative example; and identification information specification means for comparing, for each problem evoking expression, constituent elements of sentences included in the positive example solution request sentence set and the negative example solution request sentence set, and specifying a constituent element characterizing the positive example solution request sentence set and a constituent element characterizing the negative example solution request sentence set respectively as positive example identification information and negative example identification information.05-16-2013
20130124438METHOD OF RECOGNIZING PATTERNS BASED ON MARKOV CHAIN HIDDEN CONDITIONAL RANDOM FIELD MODEL - Provided is a method of recognizing patterns based on a hidden conditional random fields model to which full-Gaussian covariance has been applied. The method includes dividing a training input signal and outputting a frame sequence, extracting a feature vector from the frame sequence, calculating a parameter through a conditional random fields model to which Gaussian covariance has been applied using the feature vector, receiving, by the hidden conditional random fields model to which the parameter has been applied, a feature vector extracted from a test input signal measured for an actual pattern to infer a label indicating the actual pattern, and proposing a method of calculating gradient values for a conditional probability vector, a transition probability vector, a Gaussian mixture weight, a mean of Gaussian distributions, and covariance of the Gaussian distributions, as an analysis method.05-16-2013
20130124437SOCIAL MEDIA USER RECOMMENDATION SYSTEM AND METHOD - Each user is represented by a mixture of topics, e.g., one or more topics, and a probability of interest in each topic in the mixture, and given the target user, one or more other users can be recommended, each user that is recommended to the target user is determined to have a topical interest similarity with the target user, e.g., the target user's interest in one or more topics of the mixtures of topics is determined to be similar to a recommended interest in the one or more topics of the mixture of topics. The target user and the one or more recommended users can be said to have similar topical interests. The target user can use the user recommendation to establish an interactive dialogue, for example, with one or more users identified in the user recommendation.05-16-2013
20130124436Profiling Energy Consumption - Embodiments for detecting anomalous consumption of energy are provided. Information associated with energy consumption over a designated period of time is received. A threshold value is received. A classifier based on an Auto-Regressive Moving Average model is applied to the information and a result representing the likelihood of an attack is determined. The result is then analyzed to determine if it attained a threshold value. The information is then classified as indicating an attack. Additionally, embodiments for utilizing machine learning to train a classifier using training data to develop parameters for the auto-regressive moving average model are provided. Further, embodiments for evaluating the effectiveness of the parameters used in the Auto-Regressive Moving Average model to classify data are provided.05-16-2013
20130132310Method and System for Evaluating the Class of a Test Datum in a Large-Dimension Data Space - A method and a system for evaluating the class of a test datum in an original metric space, each datum belonging to at least one class grouping a plurality of data, includes a step of graphical representation of the spatial organization of a set of learning data of the original space in a representation metric space, a conjoint membership level indicating if any two data from the learning set belong to the same class. The method also includes a step of relating the test datum to the projections of the data from the learning set, the most probable class of the test datum being the class of the projections of the data from the learning set related to the test datum. Application: assistance with decision-making in discrimination, shape recognition.05-23-2013
20130132308Enhanced DeepQA in a Medical Environment - A DeepQA engine is enhanced to provide a digital medical investigation tool which assists a medical professional in researching potential causes of a set of patient conditions, including clues, facts and factoids about the patient. The DeepQA engine provides one or more answers to a natural language question with confidence levels for each answer. If a confidence level falls below a threshold, the enhanced DeepQA engine performs a crowd sourcing operation to gather additional information from one or more domain experts. The domain expert responses are provided to the medical professional, and are learned by the enhanced DeepQA system to provide for better research of similar patient conditions in future queries.05-23-2013
20080208774ONTOLOGY SYSTEM FOR CONTEXT-AWARE, METHOD THEREOF, AND RECORDING MEDIUM STORING THE SAME - Provided are an ontology system, a method for managing the ontology system, and a recording medium storing the same. The ontology system includes: a context broker unit for receiving context information from a sensing device and verifying a validity of the received context information; a context managing unit for controlling to generate an ontology structure by transforming the verified context information from the context broker unit to ontology web language (OWL) data and processing the OWL data; a rule-based inference engine unit for transforming the processed context information from the context managing unit to semantic web rule language (SWRL) data and processing the SWRL data through an inference process; a learning managing unit for processing the processed context information from the context managing unit through learning; and a database for storing the context information processed at the context managing unit, the rule-based engine unit, and the learning managing unit.08-28-2008
20110213739NON-INTRUSIVE LOAD MONITORING SYSTEM AND METHOD - In accordance with one embodiment, a system for non-intrusive load monitoring includes an output device, a data storage device including program instructions stored therein, a sensing device operably connected to a common source for a plurality of electrical devices, and an estimator operably connected to the output device, the data storage device, and the sensing device, the estimator configured to execute the program instructions to obtain data associated with a sensed state of the common source from the sensing device, obtain at least one model of each of the plurality of electrical devices, solve a Mixed Integer Programming problem for the at least one models over a fixed time horizon using the obtained data to determine a combination of operational stages of the plurality of electrical devices, and store operational data based on the solved Mixed Integer Programming problem.09-01-2011
20080201281PARALLEL SUPPORT VECTOR METHOD AND APPARATUS - Disclosed is an improved technique for training a support vector machine using a distributed architecture. A training data set is divided into subsets, and the subsets are optimized in a first level of optimizations, with each optimization generating a support vector set. The support vector sets output from the first level optimizations are then combined and used as input to a second level of optimizations. This hierarchical processing continues for multiple levels, with the output of each prior level being fed into the next level of optimizations. In order to guarantee a global optimal solution, a final set of support vectors from a final level of optimization processing may be fed back into the first level of the optimization cascade so that the results may be processed along with each of the training data subsets. This feedback may continue in multiple iterations until the same final support vector set is generated during two sequential iterations through the cascade, thereby guaranteeing that the solution has converged to the global optimal solution. In various embodiments, various combinations of inputs may be used by the various optimizations. The individual optimizations may be processed in parallel.08-21-2008
20080201279METHOD AND APPARATUS FOR AUTOMATICALLY STRUCTURING FREE FORM HETERGENEOUS DATA - Techniques are provided for automatically structuring free form heterogeneous data. In one aspect of the invention, the techniques include obtaining free form heterogeneous data, segmenting the free form heterogeneous data into one or more units, automatically labeling the one or more units based on one or more machine learning techniques, wherein each unit is associated with a label indicating an information type, and structuring the one or more labeled units in a format to facilitate one or more operations that use at least a portion of the labeled units, e.g., information technology (IT) operations.08-21-2008
20080201278Method and Apparatus for Automatic Online Detection and Classification of Anomalous Objects in a Data Stream - The invention is concerned with a method for automatic online detection and classification of anomalous objects in a data stream, especially comprising datasets and/or signals, wherein a) the detection of at least one incoming data stream containing normal and anomalous objects, b) automatic construction of a geometric representation of normality the incoming objects of the data stream at a time t08-21-2008
20110225108TEMPORAL MEMORY USING SPARSE DISTRIBUTED REPRESENTATION - A processing node in a temporal memory system includes a spatial pooler and a sequence processor. The spatial pooler generates a spatial pooler signal representing similarity between received spatial patterns in an input signal and stored co-occurrence patterns. The spatial pooler signal is represented by a combination of elements that are active or inactive. Each co-occurrence pattern is mapped to different subsets of elements of an input signal. The spatial pooler signal is fed to a sequence processor receiving and processed to learn, recognize and predict temporal sequences in the input signal. The sequence processor includes one or more columns, each column including one or more cells. A subset of columns may be selected by the spatial pooler signal, causing one or more cells in these columns to activate.09-15-2011
20100287128Anomaly Detection for Link-State Routing Protocols - Disclosed herein is an anomaly detection method for link-state routing protocols, a link-state routing protocol providing for link-state update (LSU) messages to be exchanged between nodes in a packet-based network, wherein each link-state update message includes link-state advertisement (LSA) message(s) each having a respective header. The method comprises monitoring the link-state advertisement messages exchanged in the network, extracting and forming respective feature vectors with the values in the fields of the headers of the monitored link-state advertisement messages, and detecting an anomaly related to routing based on the feature vectors. In particular, detecting an anomaly related to routing includes feeding the feature vectors to a machine learning system, conveniently a one-class classifier, preferably a one-class support vector machine (OC-SVM).11-11-2010
20100287127SELF-LEARNING ENGINE FOR THE REFINEMENT AND OPTIMIZATION OF SURGICAL SETTINGS - The present invention pertains to a system (or engine) that monitors a system's performance during a surgery, analyzes that performance, and makes recommendations to the user/surgeon for changes in his settings and/or programs that will result in more effective and time-efficient surgeries. Further, the system may comprise one or more components, including, but not limited to, a user preference filter, a surgical circumstances filter, a surgical instrument, a real time data collection module, and an analysis module.11-11-2010
20100287126BATTERY LEARNING SYSTEM - A fuel battery system is comprised of a power source circuit, a rotating electrical machine that is a load, a memory device and a control unit. Here, a battery learning system corresponds to an arrangement including a fuel battery that is a structural component of the power source circuit, a high frequency signal source, an electric current detection means, a voltage detection means, the memory device and a battery learning part that is a structural element of the control unit. An impedance value can be obtained from alternating current components of respective detecting values of the electric current detection means and the voltage detection means. The battery learning unit has an I-V characteristic curve learning module that learns an I-V characteristic curve and a learning prohibition judgment module that judges whether or not an acquiring interval of the impedance value is over a predetermined threshold interval set in advance, and prohibits learning if the former is over the latter.11-11-2010
20100287125INFORMATION PROCESSING UNIT, INFORMATION PROCESSING METHOD, AND PROGRAM - The present invention relates to an information processing unit, an information processing method, and a program that can allow two-class classification to be correctly performed based on the outputs from two or more classifiers.11-11-2010
20100287124HIERARCHICAL TEMPORAL MEMORY UTILIZING NANOTECHNOLOGY - Methods and systems are presented for constructing biological-scale hierarchically structured cortical statistical memory systems using currently available fabrication technology and meta-stable switching devices. Learning content-addressable memory and statistical random access memory circuits are detailed. Additionally, local and global signal modulation of bottom-up and top-down processing for the initiation and direction of behavior is disclosed.11-11-2010
20110238606KERNEL REGRESSION SYSTEM, METHOD, AND PROGRAM - In training data, a similarity matrix is generated for each of types of data corresponding to different kernels, and graph Laplacians are formed individually from the similarity matrices. An entire graph Laplacian is defined as linear combination of the individual graph Laplacians with coupling constants. Observation variables and latent variables associated therewith are assumed to form normal distributions, and the coupling constants are assumed to form a gamma distribution. Then, on the basis of a variational Bayesian method, a variance of the observation variables and the coupling constants can be figured out with a reasonable computational cost. Once the variance of the observation variables and the coupling constants are figured out, a predictive distribution for any input data can be figured out by means of a Laplace approximation.09-29-2011
20130151440METHOD OF REGENERATING DIFFRACTION SIGNALS FOR OPTICAL METROLOGY SYSTEMS - Provided is a method for enhancing accuracy of an optical metrology system that includes a metrology tool, an optical metrology model, and a profile extraction algorithm. The optical metrology model includes a model of the metrology tool and a profile model of the sample structure, the profile model having profile parameters. A library comprising Jones and/or Mueller matrices and/or components (JMMOC) and corresponding profile parameters is generated using ray tracing and a selected range of beam propagation parameters. An original simulated diffraction signal is calculated using the optical metrology model. A regenerated simulated diffraction signal is obtained using the regenerated JMMOC, integrated for all the rays of the optical metrology model. If an error and precision criteria for the regenerated simulated diffraction signal compared to the original simulated diffraction signal are met, one or more profile parameters are determined from the best match regenerated simulated diffraction signal.06-13-2013
20130151441MULTI-TASK LEARNING USING BAYESIAN MODEL WITH ENFORCED SPARSITY AND LEVERAGING OF TASK CORRELATIONS - Multi-task regression or classification includes optimizing parameters of a Bayesian model representing relationships between D features and P tasks, where D≧1 and P≧1, respective to training data comprising sets of values for the D features annotated with values for the P tasks. The Bayesian model includes a matrix-variate prior having features and tasks dimensions of dimensionality D and P respectively. The matrix-variate prior is partitioned into a plurality of blocks, and the optimizing of parameters of the Bayesian model includes inferring prior distributions for the blocks of the matrix-variate prior that induce sparseness of the plurality of blocks. Values of the P tasks are predicted for a set of input values for the D features using the optimized Bayesian model. The optimizing also includes decomposing the matrix-variate prior into a product of matrices including a matrix of reduced rank in the tasks dimension that encodes correlations between tasks.06-13-2013
20130151442METHOD FOR LEARNING TASK SKILL AND ROBOT USING THEREOF - Provided are a method for learning task skill and a robot using the same. The modeling method for learning a task skill includes: receiving training data for a task to be performed by a learning engine; dividing, by the learning engine, the received training data into segments by using a geometric property of a predetermined probabilistic model; and learning, by the learning engine, a basis skill for the divided segments by modeling each divided segment.06-13-2013
20130151443Systems and methods for performing contextual classification using supervised and unsupervised training - Computerized systems and methods are disclosed for performing contextual classification of objects using supervised and unsupervised training. In accordance with one implementation, content reviewers may review training objects and submit supervised training data for preprocessing and analysis. The supervised training data may be preprocessed to identify key terms and phrases, such as by stemming, tokenization, or n-gram analysis, and form vectorized objects. The vectorized objects may be used to train one or more models for subsequent classification of objects. In certain implementations, preprocessing or training, among other steps, may be performed in parallel over multiple machines to improve efficiency. The disclosed systems and methods may be used in a wide variety of applications, such as article classification and content moderation.06-13-2013
20130151444METHODS AND APPARATUS FOR UTILISING SOLUTIONS TO SAT PROBLEMS - Computer implemented method to indicate whether a CNF sentence representing a physical system is satisfiable. The method includes structuring a search tree based upon received data representing the CNF sentence. The search tree includes a root node and a plurality of other nodes. The method includes causing the computer to use a search to visit nodes using a decision heuristic at each node to determine which of the branches of the search tree to explore from that node, determining which nodes lie on the solution path, modifying the decision heuristic according to the analysis, generating a trained decision heuristic, and using the trained decision heuristic to process CNF sentences to determine whether those CNF sentences are satisfiable. A shortest path through the search tree provides a solution path and the heuristic can be trained with a set of training instances.06-13-2013
20130151445Method and System for Survival of Data Plane Through a Total Control Plane Failure - A system and method for retaining routes in a control plane learned by an inter-domain routing protocol in the event of a connectivity failure between routers. Routers are classified as either route reflectors or originators. A determination is made whether the connectivity failure occurred between a route reflector and an originator, two originators, or two route reflectors. A determination is then made whether to propagate a withdrawal of learned routes based on whether the connectivity failure occurred between a route reflector and an originator, two originators, or two route reflectors. A withdrawal of learned routes is propagated to neighboring routers if the connectivity failure occurred between two originators, or between a route reflector and an originator that is inaccessible via an intra-domain routing protocol. No withdrawal of learned routes is propagated if the connectivity failure occurred between two route reflectors, or between a route reflector and an originator that is accessible via an intra-domain routing protocol.06-13-2013
20130151446METHOD AND SYSTEM FOR DETERMINING THE ACCURACY OF DNA BASE IDENTIFICATIONS - Embodiments disclosed herein relate to a method and system for determining the accuracy of DNA base identifications, based at least partly on sampling characteristics of subsets within training data sets.06-13-2013
20130151447METHOD AND APPARATUS FOR SELF-LEARNING AND SELF-IMPROVING A SEMICONDUCTOR MANUFACTURING TOOL - Performance of a manufacturing tool is optimized. Optimization relies on recipe drifting and generation of knowledge that capture relationships among product output metrics and input material measurement(s) and recipe parameters. Optimized recipe parameters are extracted from a basis of learned functions that predict output metrics for a current state of the manufacturing tool and measurements of input material(s). Drifting and learning are related and lead to dynamic optimization of tool performance, which enables optimized output from the manufacturing tool as the operation conditions of the tool changes. Features of recipe drifting and associated learning can be autonomously or externally configured through suitable user interfaces, which also can be drifted to optimize end-user interaction.06-13-2013
20110258153COMBINING PREDICTIVE MODELS OF FORGETTING, RELEVANCE, AND COST OF INTERRUPTION TO GUIDE AUTOMATED REMINDING - The claimed matter provides systems and/or techniques that develop or use predictive models of human forgetting to effectuate automated reminding. The system includes the use of predictive models that infer the probability that aspects of items will be forgotten, models that evaluate the relevance of recalling aspects of items in different settings, based on contextual information related to user attributes associated with the items, and models of the context-sensitive cost of interrupting users with reminders. The system can combine the probability of users forgetting aspects of an item with an assessed cost of forgetting those aspects to ascertain expected costs for not being reminded about events, compare expected costs for not being reminded with expected costs for interrupting users, and based on comparisons between expected costs for being reminded and expected costs for interrupting users regarding events, generate and deliver reminder notifications to users about items.10-20-2011
20130132311SCORE FUSION AND TRAINING DATA RECYCLING FOR VIDEO CLASSIFICATION - Multiple classifiers can be applied independently to evaluate images or video. Where there are heavily imbalanced class distributions, a local expert forest model for meta-level score fusion for event detection can be used. Performance variations of classifiers in different regions of a score space can be adapted. Multiple pairs of experts based on different partitions, or “trees,” can form a “forest,” balancing local adaptivity and over-fitting. Among ensemble learning methods, stacking with a meta-level classifier can be used to fuse an output of multiple base-level classifiers to generate a final score. A knowledge-transfer framework can reutilize the base-training data for learning the meta-level classifier. By recycling the knowledge obtained during a base-classifier-training stage, efficient use can be made of all available information, such as can be used to achieve better fusion and better overall performance.05-23-2013
20130132309Method Performed in a Computer System for Aiding the Assessment of an Influence of a User in or Interacting with a Communication System by Applying Social Network Analysis, SNA, Functions, a Computer System, Computer Program and Computer Program Product - The invention relates to a method performed in a computer system for aiding the assessment of an influence of a user in or interacting with a communication system by applying social network analysis, SNA, functions. The method comprises: obtaining two or more SNA metrics for each user of a first number of users, each SNA metric being determined by a respective SNA function; calculating a weight parameter for each one of the SNA metrics using a machine learning method, the weight parameters indicating a combination of the SNA metrics for use in the assessment of the influence of the user; and applying the estimated weight parameters to SNA metrics of a second number of users to assess a ranking in accordance with influence of users in the second number of users. The invention also relates to a computer system, computer programs, and computer program products.05-23-2013
20100306140COMBINATION CONTAMINANT SIZE AND NATURE SENSING SYSTEM AND METHOD FOR DIAGNOSING CONTAMINATION ISSUES IN FLUIDS - Systems and methods used to monitor a fluid where it is important to know the size, concentration and nature of particulates in the fluid. For example, the systems and method can be used to diagnose contamination issues in fluids such as fuel, lubrication, power transfer, heat exchange or other fluids in fluid systems, for example diesel engines or hydraulic systems, where contaminant particles in the fluids are of concern.12-02-2010
20100318481Generating Test Data - Generating test data includes: reading values occurring in at least one field of multiple records from a data source; storing profile information including statistics characterizing the values; generating a model of a probability distribution for the field based on the statistics; generating multiple test data values using the generated model such that a frequency at which a given value occurs in the test data values corresponds to a probability assigned to that given value by the model; and storing a collection of test data including the test data values in a data storage system.12-16-2010
20100318480LEARNING CONTROL SYSTEM AND LEARNING CONTROL METHOD - A learning control system according to the present invention is one which performs learning of action values of actions in an apparatus which identifies its state as one of predetermined states, and selects an action based on the obtained action values and the identified state. The learning control system includes n action value learning devices including the first to the n th learning devices which perform learning of n action values from Q12-16-2010
20100318479INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing device includes: a learning section configured to learn a state transition probability model defined by state transition probability for each action of a state making a state transition due to an action performed by an agent capable of performing action and observation probability of a predetermined observed value being observed from the state, using an action performed by the agent and an observed value observed in the agent when the agent has performed the action.12-16-2010
20100318478INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM - An information processing device includes: a calculating unit configured to calculate a current-state series candidate that is a state series for an agent capable of actions reaching the current state, based on a state transition probability model obtained by performing learning of the state transition probability model stipulated by a state transition probability that a state will be transitioned according to each of actions performed by an agent capable of actions, and an observation probability that a predetermined observation value will be observed from the state, using an action performed by the agent, and an observation value observed at the agent when the agent performs an action; and a determining unit configured to determine an action to be performed next by the agent using the current-state series candidate in accordance with a predetermined strategy.12-16-2010
20100318477FAST AND EFFICIENT NONLINEAR CLASSIFIER GENERATED FROM A TRAINED LINEAR CLASSIFIER - A classifier method comprises: projecting a set of training vectors in a vector space to a comparison space defined by a set of reference vectors using a comparison function to generate a corresponding set of projected training vectors in the comparison space; training a linear classifier on the set of projected training vectors to generate a trained linear classifier operative in the comparison space; and transforming the trained linear classifier operative in the comparison space into a trained nonlinear classifier that is operative in the vector space to classify an input vector.12-16-2010
20120284215METRIC LEARNING APPARATUS - A metric learning apparatus memorizes a learning pattern in a feature space and a category which the learning pattern belongs to, performs variable transformation of the learning pattern to a metric space by a transformation matrix, calculates a transformation matrix having a minimum loss value of a loss function in which the loss value is increased when there is a learning pattern belonging to a different category but closer than learning patterns up to k11-08-2012
20120284214SOFTWARE, DISPLAY AND COMPUTER SYSTEM FOR RUNNING AND PRESENTING IMAGES AS PART OF THERAPY FOR ENHANCING NUMERICAL COGNITION - A method of presenting training materials for training users with developmental dyscalculia or related learning difficulties includes determining a number or a numerical expression to present as part of training the user in developing internal maps to assist with overcoming a learning difficulty, whereby the user can increase a tendency to establish an internal neurological representation of numbers and numerical expression, wherein a numerical expression is a sequence of at least one number and at least one mathematical operator, generating a representation in a virtual space of an arrangement of numbers, including a number line and a representation of the number or the numerical expression, taking into account a resolution of the computer-controlled display that is to be used, and presenting to the user, using the computer-controlled display, a view of the virtual space showing the number line and the representation of the number or the numerical expression.11-08-2012
20120284213Predictive Analytical Modeling Data Selection - A system includes a computer(s) coupled to a data storage device(s) that stores a training data repository and a predictive model repository. The training data repository includes retained data samples from initial training data and from previously received data sets. The predictive model repository includes at least one updateable trained predictive model that was trained with the initial training data and retrained with the previously received data sets. A new data set is received. A richness score is assigned to each of the data samples in the set and to the retained data samples that indicates how information rich a data sample is for determining accuracy of the trained predictive model. A set of test data is selected based on ranking by richness score the retained data samples and the new data set. The trained predictive model is accuracy tested using the test data and an accuracy score determined.11-08-2012
20120284212Predictive Analytical Modeling Accuracy Assessment - A system includes a computer(s) coupled to a data storage device(s) that stores a training function repository and a predictive model repository that includes includes updateable trained predictive models each associated with an accuracy score. A series of training data sets are received, being training samples each having output data that corresponds to input data. The training data is different from initial training data that was used with training functions from the repository to train the predictive models initially. Upon receiving a first training data set included in the series and for each predictive model in the repository, the input data in the first training set is used to generate predictive output data that is compared to the output data. Based on the comparison and previous comparisons determined from the initial training data and from previously received training data sets, an updated accuracy score for each predictive model is determined.11-08-2012
20120284216Knowledge-Based Models for Data Centers - Techniques for data center analysis are provided. In one aspect, a method for modeling thermal distributions in a data center includes the following steps. Vertical temperature distribution data is obtained for a plurality of locations throughout the data center and is plotted as an s-curve, wherein the vertical temperature distribution data reflects physical conditions at each of the locations which is reflected in a shape of the s-curve. Each of the s-curves is represented with a set of parameters that characterize the shape of the s-curve, wherein the s-curve representations make up a knowledge base model of predefined s-curve types from which thermal distributions and associated physical conditions at the plurality of locations throughout the data center can be analyzed. The set of parameters that characterize the shape of the s-curve are associated with the physical conditions at the plurality of locations throughout the data center using a machine-learning model.11-08-2012
20130097108Two-Stage Multiple Kernel Learning Method - Disclosed are methods and structures of Multiple Kernel learning framed as a standard binary classification problem with additional constraints that ensure the positive definiteness of the learned kernel. Advantageously, the disclosed methods and structures permit the use of binary classification technologies to develop better performing, and more scalable Multiple Kernel Learning methods that are conceptually simpler.04-18-2013
20130159223Virtual Sensor Development - Embodiments include processes, systems, and devices for developing a virtual sensor. The virtual sensor includes one or more inference models. A decision engine utilizes an inference model associated with a mobile device to determine another inference model that is configured to accept physical sensor data from another mobile device. In this way, the virtual sensor can be developed for use with many mobile devices using initial inference models developed for a small number of mobile devices or a single mobile device. Embodiments also include methods to select mobile devices from which to request physical sensor data for virtual sensor input. Embodiments also include architectures that provide a library of virtual sensors.06-20-2013
20130159225MODEL BASED CALIBRATION OF INFERENTIAL SENSING - An inferential sensor module is incorporated into an engine simulation model. One or more parameters for the inferential sensor module are calibrated using one or more of engine measurement data and the engine simulation model. The calibration is performed such that a difference between an inferred signal predicted by the inferential sensor module and a signal measured on an engine is minimized. The inferential sensor module and the one or more calibrated parameters are loaded into an engine control unit in order to predict inferred variables.06-20-2013
20130159226Method for Screening Samples for Building Prediction Model and Computer Program Product Thereof - A method for screening samples for building a prediction model and a computer program product thereof are provided. When a set of new sample data is added to a dynamic moving window (DMW), a clustering step is performed with respect to all of the sets of sample data within the window for grouping the sets of sample data with similar properties as one group. If the number of the sets of sample data in the largest group is greater than a predetermined threshold, it means that there are too many sets of sample data with similar properties in the largest group, and the oldest sample data in the largest group can be deleted; if smaller than or equal to a predetermined threshold, it means that the sample data in the largest group are quite unique, and should be kept for building or refreshing the prediction model.06-20-2013
20130159227CLUSTERING COOKIES FOR IDENTIFYING UNIQUE MOBILE DEVICES - Embodiments are directed towards clustering cookies for identifying unique mobile devices for associating activities over a network with a given mobile device. The cookies are clustered based on a Bayes Factor similarity model that is trained from cookie features of known mobile devices. The clusters may be used to determine the number of unique mobile devices that access a website. The clusters may also be used to provide targeted content to each unique mobile device.06-20-2013
20130159221Systems for Monitoring Computer Resources - One embodiment of a system of the present invention for monitoring computer resources includes means for retrieving a set of resource-metric records for a predetermined time interval, means for forming a first mathematical matrix containing metric's values arranged on date-time and resource-metric axes, means for creating a second mathematical matrix containing features and a third mathematical matrix containing weights, means for building a feature relationship tree, means for generating a predicted value for the resource-metric identifier, means for determining a variance between predicted value and metric's value, and means for triggering an alert if the variance exceeds a predetermined alert threshold.06-20-2013
20130185231PREDICTING DIAGNOSIS OF A PATIENT - Method, system, and computer program product are provided for predicting diagnosis of a patient performed by a computerized device. The method may include: modeling data from a group of successfully diagnosed patients, wherein the data is modeled as treatment paths of patients including referrals to medical practitioners; and predicting diagnosis for a current patient by comparing a treatment path of the current patient with the modeled treatment paths of successfully diagnosed patients, including calculating a probability of a given diagnosis from the modeled treatment paths. The method may include: defining a set of medical entities including medical practitioners to which a patient has been referred; and gathering treatment paths of successfully diagnosed patients, wherein the treatment path links medical entities in a directional route. Predicting diagnosis for a current patient may use the modeled data to calculate the probability of each model instance for each diagnosis and choosing the model instance of the diagnosis that maximizes the treatment path probability.07-18-2013
20130185230MACHINE-LEARNING BASED CLASSIFICATION OF USER ACCOUNTS BASED ON EMAIL ADDRESSES AND OTHER ACCOUNT INFORMATION - A trust level of an account is determined at least partly based on a degree of the memorability of an email address associated with the account. Additional features such as those based on the domain of the email address and those from the additional information such as name, phone number, and address associated with the account may also be used to determine the trust level of the account. A machine learning process may be used to learn a classification model based on one or more features that distinguish a malicious account from a benign account from training data. The classification model is used to determine a trust level of the account, and/or if the account is malicious or benign, and may be continuously improved by incrementally adapting or improving the model with new accounts.07-18-2013
20130185233SYSTEM AND METHOD FOR LEARNING POSE CLASSIFIER BASED ON DISTRIBUTED LEARNING ARCHITECTURE - A system and method for learning a pose classifier based on a distributed learning architecture. A pose classifier learning system may include an input unit to receive an input of a plurality of pieces of learning data, and a plurality of pose classifier learning devices to receive an input of a plurality of learning data sets including the plurality of pieces of learning data, and to learn each pose classifier. The pose classifier learning devices may share learning information in each stage, using a distributed/parallel framework.07-18-2013
20110282811Method and Data Processing System for Automatic Identification, Processing, Interpretation and Evaluation of Objects in the Form of Digital Data, Especially Unknown Objects - The invention relates to methods and data processing systems for the automatic identification, processing, interpretation and evaluation of objects in the form of digital data, especially unknown objects. Said methods and systems are characterised in that objects which cannot be associated with any known model are entered into a case database as unknown objects. They are then available for automatic interpretation and evaluation by means of a similarity-based method. Said unknown objects can lead to new models. The model database is thereby continuously enlarged, existing models refined, and new models learned. The model data-base can be organised evenly or hierarchically in statistical models representing higher classes and lower classes.11-17-2011
20080281766Time Machine Software - A method and system for creating human robots with psychic abilities, as well as enabling a human robot to access information in a time machine to predict the future accurately and realistically. The present invention provides a robot with the ability to accomplish tasks quickly and accurately without using any time. This permits a robot to cure cancer, fight a war, write software, read a book, learn to drive a car, draw a picture or solve a complex math problem in less than one second.11-13-2008
20100057650CHEMICAL REACTION-TYPE METAHEURISTIC - Subject matter disclosed herein relates to various embodiments of a chemical reaction-type metaheuristic. According to an embodiment, solutions to an objective function can be determined by iteratively searching for a minimum energy state of one or more interactions of molecules in a chemical reaction. The molecules in the chemical reaction can be assigned to represent the possible outcomes of the objective function. In a specific embodiment, the interactions of the molecules can modeled as on-wall ineffective collisions, decompositions, inter-molecular ineffective collisions, and synthesis. The type of interaction can affect where the next molecular structure is searched.03-04-2010
20110320387Graph-based transfer learning - Transfer learning is the task of leveraging the information from labeled examples in some domains to predict the labels for examples in another domain. It finds abundant practical applications, such as sentiment prediction, image classification and network intrusion detection. A graph-based transfer learning framework propagates label information from a source domain to a target domain via the example-feature-example tripartite graph, and puts more emphasis on the labeled examples from the target domain via the example-example bipartite graph. An iterative algorithm renders the framework scalable to large-scale applications. The framework propagates the label information to both features irrelevant to the source domain and unlabeled examples in the target domain via common features in a principled way.12-29-2011
20120011085Systems And Methods For Identifying And Notifying Users of Electronic Content Based on Biometric Recognition - Systems and methods are disclosed for manipulating electronic multimedia content to a user. One method includes generating a plurality of biometric models, each biometric model corresponding to one of a plurality of people; receiving electronic media content over a network; extracting image or audio data from the electronic media content; detecting biometric information in the image or audio data; and calculating a probability of the electronic media content involving one of the plurality of people, based on the biometric information and the plurality of biometric models.01-12-2012
20120011084SEMANTIC ENTITY MANIPULATION USING INPUT-OUTPUT EXAMPLES - Semantic entity manipulation technique embodiments are presented that generate a probabilistic program capable of manipulating character strings representing semantic entities based on input-output examples. The program can then be used to produce a desired output consistent with the input-output examples from inputs of a type included in the examples. The probabilistic program is generated based on the output of parsing, transform and formatting modules. The parsing module employs a probabilistic approach to parsing the input-output examples. The transform module identifies a weighted set of transforms that are capable of producing the output item from the input items of an input-output example to a likelihood specified by their assigned weight. The formatting module generates formatting instructions that transform selected output parts into a form specified by the output items in the input-output examples.01-12-2012
20120016821INFORMATION PROCESSING DEVICE, INFORMATION PROCESSING METHOD, AND PROGRAM - There is provided a method including inputting a plurality of symbol strings and attribute information desired to be extracted from each symbol string; selecting a plurality of functions from a predetermined function group including a function for converting a symbol string into a numerical value, and generating a plurality of feature quantity functions for outputting a feature quantity from the symbol string by combining the plurality of functions; inputting each symbol string to each feature quantity function, and calculating a feature quantity corresponding to each symbol string; executing machine learning using the attribute information corresponding to each symbol string and the feature quantity corresponding to each symbol string, and generating an estimation function for estimating the attribute information from the feature quantity; and outputting the feature quantity functions and the estimation function.01-19-2012
20120023047Method for a Pattern Discovery and Recognition - A method is for a pattern discovery and recognition, wherein a first sequence comprising first sequence symbols relating to a concept and a tag associated to the first sequence are received, transition probability matrices are obtained from transition frequency matrices representing the frequency data of the occurrences of the transitions between the first sequence symbols at different distances in the first sequence, and the transition probability matrices for each tag and each distance are learnt for obtaining an activation function determining the concept occurring in a second sequence. A computer program product and an apparatus are for executing the pattern discovery and recognition method.01-26-2012
20120023046Deducing Shadow User Profiles For Ad Campaigns - A method and a system are provided for deducing shadow user profile attributes for ad campaigns aimed at target users. In one example, the system extracts tagged data from source data. The tagged data includes label information associated with an actual profile for a user. The tagged data is associated with the user. The system prepares the tagged data by splitting the tagged data into datasets, including at least training data and test data. The system generates one or more individual models based on the tagged data, wherein the one or more individual models provide the ability to deduce attributes of a profile for the user. The system then generates a composite model based on the individual models. The composite model includes a combination of the individual models that are associated with the user. The system may charge a premium for ad campaigns that are aimed at target users who are each assigned one or more shadow profile attribute values. The system may determine the premium based on the confidence level with which the one or more attribute values fits to the one or more users. The system is applicable to both display advertising and sponsored search advertising.01-26-2012
20120023043Estimating Probabilities of Events in Sponsored Search Using Adaptive Models - A machine-learning method for estimating probability of a click event in online advertising systems by computing and comparing an aggregated predictive model (a global model) and one or more data-wise sliced predictive models (local models). The method comprises receiving training data having a plurality of features stored in a feature set and constructing a global predictive model that estimates the probability of a click event for the processed feature set. Then, partitioning the global predictive model into one or more data-wise sliced training sets for training a local model from each of the data-wise slices, and then determining whether a particular local model estimates probability of click event for the feature set better than the global model. A given feature set may be collected from historical data, and may comprise a feature vector for a plurality of query-advertisement pairs and a corresponding indicator that represents a click on the advertisement.01-26-2012
20120030156COMPUTER-IMPLEMENTED METHOD, CLINICAL DECISION SUPPORT SYSTEM, AND COMPUTER-READABLE NON-TRANSITORY STORAGE MEDIUM FOR CREATING A CARE PLAN - A computer-implemented method for creating a care plan for a patient, the method comprising the steps of (a) receiving an intervention goal, (b) creating a behavioral determinants list using the intervention goal, (c) assigning a ranking to each behavioral determinant in the list of behavioral determinates using patient data descriptive of the patient, (d) receiving a subset of the behavioral determinant list, wherein the subset is determined by using the ranking of each behavioral determinant in the list of behavioral determinates; and (f) creating the care plan using the subset.02-02-2012
20120030155MODEL GENERATING DEVICE AND MODEL GENERATING METHOD - A model generating device acquires event information including a time when execution of an event is started, a time when execution of the event is finished, and the type of the event. The model generating device assumes model candidates each showing a relationship between an event and another event triggered by the former event on the basis of the acquired event information, and makes a combination of the assumed model candidates. The model generating device searches the assumed model candidates to find a candidate that matches an existing model. The model generating device withdraws the model candidate searched for from the combination of the model candidates, and withdraws a candidate that is to be withdrawn from the combination of the model candidates in association with withdrawal of the model candidate found as a result of the search, thereby updating the combination of the model candidates. The model generating device thereafter generates a new model on the basis of the updated combination of the model candidates.02-02-2012
20120030154ESTIMATING A STATE OF AT LEAST ONE TARGET - A method of estimating a state of at least one target. The method includes obtaining at least one target measurement from a first sensor, and applying a Gaussian Process technique to a target measurement to obtain an updated target measurement.02-02-2012
20120030153SEMICONDUCTOR SYSTEM AND DATA TRAINING METHOD THEREOF - A semiconductor system includes a semiconductor memory configured to determine whether an error has occurred in a data pattern and generate an error signal, and a memory controller configured to provide the data pattern to the semiconductor memory and perform data training with respect to the semiconductor memory using the error signal.02-02-2012
20120030152RANKING ENTITY FACETS USING USER-CLICK FEEDBACK - Example methods, apparatuses, or articles of manufacture are disclosed that may be implemented using one or more computing devices to facilitate or otherwise support one or more processes or operations associated with ranking entity facets using user-click feedback.02-02-2012
20120030151METHOD AND SYSTEM FOR ASSESSING DATA CLASSIFICATION QUALITY - Production data classified from a data source, such as a plurality of handprinted forms, is compared to provisional truth data independently classified from the same data source for constructing master truth data. The production data is compared to the master truth data for evaluating the quality with which the production data was classified.02-02-2012
20120030150Hybrid Learning Component for Link State Routing Protocols - In a network that executes a link state routing protocol, a network node receives periodic disseminations of link state information from other network nodes. The link state information includes neighboring node identity and link cost metrics. The network node calculates the initial routing paths based on the received link state information by using a link state routing algorithm. It then adapts the calculated path based on both the current link state information and past link state information through a reinforcement learning process. The network node then selects a routing path to each destination node based on the adaptation and updates the routing table accordingly.02-02-2012
20130198119APPLICATION OF MACHINE LEARNED BAYESIAN NETWORKS TO DETECTION OF ANOMALIES IN COMPLEX SYSTEMS - According to one embodiment, in response to a set of data for anomaly detection, a Bayesian belief network (BBN) model is applied to the data set, including for each of a plurality of features of the BBN model, performing an estimate using known observed values associated with remaining features to generate a posterior probability for the corresponding feature. A scoring operation is performed using a predetermined scoring algorithm on posterior probabilities of all of the features to generate a similarity score, wherein the similarity score represents a degree to which a given event represented by the data set is novel relative to historical events represented by the BBN model.08-01-2013
20130198120SYSTEM AND METHOD FOR PROFESSIONAL CONTINUING EDUCATION DERIVED BUSINESS INTELLIGENCE ANALYTICS - The present disclosure relates to non-linear analytics engine derived business intelligence. More particularly, the present disclosure describes methods and systems that use content associated with a medical professional continuing education event as a data source for a non-linear analytics engine. The content, which relates to the content creation, content delivery, follow-ups, evaluations, attendee interactions, and administrative tasks associated with medical professional continuing education event, is extracted from a learning management system and subsequently transmitted to the non-linear analytics engine to create business intelligence.08-01-2013
20130198113METHOD AND TECHNIQUE TO CREATE SINGLE INTELLIGENT COLLABORATION PLATFORM SPANNING ACROSS WEB, MOBILE AND CLOUD - A method that knits together and logically sequences diverse services such as, but not limited to, social networks, financial services, news feeds, email services, calendar services, analytical platforms and Business-to-consumer (B2C) services to create a state full cohesive end-to-end user experience on a single intelligent collaboration platform spanning across web, mobile and cloud.08-01-2013
20130198114Classifying Activity Using Probabilistic Models - A method, an apparatus and an article of manufacture for classifying customer activity in an automated customer support system. The method includes obtaining input from the automated customer support system, wherein the input comprises an observable measurement of customer activity in the automated customer support system, computing a probability that the input corresponds to one of one or more probabilistic models, and using the computed probability to classify the customer activity in the automated customer support system by considering the probabilistic model corresponding to a highest computed probability.08-01-2013
20130198115CLUSTERING CROWDSOURCED DATA TO CREATE AND APPLY DATA INPUT MODELS - The collection and clustering of data input characteristics from a plurality of computing devices is provided. The clustered data input characteristics define user groups to which users are assigned. Input models such as language models and touch models are created for, and distributed to, each of the user groups based on the data input characteristics of the users assigned thereto. For example, an input model may be selected for a computing device based on a current context of the computing device. The selected input model is applied to the computing device during the current context to alter the interpretation of input received from the user via the computing device.08-01-2013
20130198116LEVERAGING USER-TO-TOOL INTERACTIONS TO AUTOMATICALLY ANALYZE DEFECTS IN IT SERVICES DELIVERY - An approach is presented for identifying related problem tickets in an information technology (IT) environment. User interactions with a computer program are stored. The user interactions include inputs to the computer program to search for problem tickets issued in the IT environment that have the same characteristics. One or more user interaction patterns within the user interactions are recognized. A user interaction pattern of the one or more user interaction patterns is selected based on an evaluation of effectiveness of each of the one or more user interaction patterns. Based on the user interaction pattern, a rule is generated for determining which problem tickets in the IT environment share a common characteristic. The rule is applied to additional problem tickets issued in the IT environment to identify which of the additional problem tickets share the common characteristic.08-01-2013
20130198118ANNOTATION OF A BIOLOGICAL SEQUENCE - A computer-implemented method for annotation of a biological sequence, comprising: applying a classifier to determine a label for the first segment of a first biological sequence of a first species based on an estimated relationship between second segments in a training set and known labels of the second segments in the training set. The classifier is trained using the training set to estimate the relationship, and the second segments are of a second biological sequence of a second species that is different to, or a variant of, the first species. This disclosure also concerns a computer program and a computer system for annotation of a biological sequence.08-01-2013
20130198117SYSTEMS AND METHODS FOR SEMANTIC DATA INTEGRATION - Embodiments of the present invention relate to a system for data integration and information retrieval by bringing semantically related data together for a given context. As described, the integration of data may include the building of an ontology, the mapping of one or more processes, semantic maps and concept dictionaries in the ontology to one or more data sources, tagging the data sources in accordance with the ontology, providing a query interface for accepting an input query from a user, the mapping of the input query to one or more concepts in the ontology, and deriving one or more subqueries thereby, and the querying of data sources in accordance with the composed one or more subqueries, wherein the data sources queried are tagged with one or more concepts from the ontology. Additionally, the tracking of data across data sources in accordance with a defined data value chain is disclosed.08-01-2013
20130204813SELF-LEARNING, CONTEXT AWARE VIRTUAL ASSISTANTS, SYSTEMS AND METHODS - A virtual assistant learning system is presented. A monitoring device, a cell phone for example, observes user interactions with an environment by acquiring sensor data. The monitoring device uses the sensor data to identify the interactions, which in turn is provided to an inference engine. The inference engine leverages the interaction data and previously stored knowledge elements about the user to determine if the interaction exhibits one or more user preferences. The inference engine can use the preferences and interactions to construct queries targeting search engines to seek out possible future interactions that might be of interest to the user.08-08-2013
20130204812METHOD FOR COMPUTER-AIDED CLOSED-LOOP AND/OR OPEN-LOOP CONTROL OF A TECHNICAL SYSTEM - A method for computer-aided closed and/or open-loop control of a technical system is provided. A first value of an output quantity is predicted on a data-based model at a current point in time. A second value of the output quantity is determined from an analytical model. The state of the technical system at the current point is assigned a confidence score in the correctness of prediction of the data-based model. A third value of the output quantity is determined from the first and second value as a function of the confidence score for controlling the technical system. A suitable value for the output quantity can be derived from the analytical model even for regions of the technical system in which the quality of prediction of the data-based model is low because of a small set of training data. The technical systems can be turbines, such as gas turbines.08-08-2013
20130204808Fault Prediction of Monitored Assets - Systems and methods for fault prediction are described to reduce equipment failure by effectively monitoring equipment, removing anomalous data, and reducing false alarms. Such systems and methods can be used to receive monitoring data, extract information from the data, and combine extracted information for establishing prediction models. Additionally, fault probabilities may be quantified and faults may be predicted based on the probabilities.08-08-2013
20130204809ESTIMATION OF PREDICTIVE ACCURACY GAINS FROM ADDED FEATURES - Various technologies described herein pertain to estimating predictive accuracy gain of a potential feature added to a set of features, wherein an existing predictor is trained on the set of features. Outputs of the existing predictor for instances in a dataset can be retrieved from a data store. Moreover, a predictive accuracy gain estimate of a potential feature added to the set of features can be measured as a function of the outputs of the existing predictor for the instances in the dataset. The predictive accuracy gain estimate can be measured without training an updated predictor on the set of features augmented by the potential feature.08-08-2013
20130204811OPTIMIZED QUERY GENERATING DEVICE AND METHOD, AND DISCRIMINANT MODEL LEARNING METHOD - To provide an optimized query generating device capable of generating an optimized query to be given with domain knowledge when generating a discriminant model on which the domain knowledge indicating user's knowledge or analysis intention for a model is reflected.08-08-2013
20130204810DISCRIMINANT MODEL LEARNING DEVICE, METHOD AND PROGRAM - To provide a discriminant model learning device capable of efficiently learning a discriminant model on which domain knowledge indicating user's knowledge or analysis intention for a model is reflected while keeping fitting to data.08-08-2013
20120078820MIME Technology: Using EEG brainwave data obtained via a dry or wet sensor (wired or non-wired) device to direct or otherwise influence the outcome of any media file, including but not limited to a video, video advertisement or movie file. - A method, system and process using brainwave electro-encephalographic data from any sensor to direct or influence the outcome of any media file including but not limited to video, video advertisement or movie file sequences via a media player. It uses algorithm data based upon the brainwave readings and conveys them to a media player which refers to a text based computer-scripting language file to direct the outcome or sequence of scenes in a media file playable through various platforms, including but not limited to personal computers, television, Digital Video Disc, cinema screens, handheld devices and video consoles. The computer-scripting file uses pre-defined alternative outcomes in accordance with the user's state of mind as measured by the EEG device and directs the player to play the appropriate scene or time code within a video, video advertisement or movie file, thus enabling a user to influence the progress/outcome of a file.03-29-2012
20120084239Methods and Systems for Constructing Bayesian Belief Networks - Methods and systems are described for simplifying a causal influence model that describes influence of parent nodes X04-05-2012
20120084238System and Method to Enable Training a Machine Learning Network in the Presence of Weak or Absent Training Exemplars - Described is a system and method for training a machine learning network. The method comprises initializing at least one of nodes in a machine learning network and connections between the nodes to a predetermined strength value, wherein the nodes represent factors determining an output of the network, providing a first set of questions to a plurality of users, the first set of questions relating to at least one of the factors, receiving at least one of choices and guesstimates from the users in response to the first set of questions and adjusting the predetermined strength value as a function of the choices/guesstimates. The real and simulated examples presented demonstrate that synthetic training sets derived from expert or non-expert human guesstimates can replace or augment training data sets comprised of actual training exemplars that are too limited in size, scope, or quality to otherwise generate accurate predictions.04-05-2012
20120084235STRUCTURED PREDICTION MODEL LEARNING APPARATUS, METHOD, PROGRAM, AND RECORDING MEDIUM - A structured prediction model learning apparatus, method, program, and recording medium maintain prediction performance with a smaller amount of memory. An auxiliary model is introduced by defining the auxiliary model parameter set θ04-05-2012
20130212049Machine Evolutionary Behavior by Embedded Collaborative Learning Engine (eCLE) - This patent develops and demonstrates the technology required for constructing machine evolutionary behavior within systems to enable evolving learning capability for autonomous recognition of new emerging behaviors. A purpose of this technology is to provide a formal methodology and implementation for adding new knowledge, which results from the automated recognition of new patterns (behaviors) within systems. Key characteristic of the “Machine Evolutionary Behavior by Embedded Collaborative Learning engine” consist on operating with an ensemble of learning paradigms, which when instantiated work in a collaborative way. The resulting framework compiles the inherent advantages of the involved methods, but also a synergetic behavior is obtained when working in a collaborative fashion.08-15-2013
20130212048METHOD OF performing REAL-TIME CORRECTION OF A WATER STAGE FORECAST - A method of performing real-time correction of a water stage forecast includes obtaining at least one predicted water stage of at least one time and a predicted water stage of a next time after the at least one time; obtaining at least one observed water stage of the at least one time; generating a system error of the water stage forecast according to the at least one observed water stage, the at least one predicted water stage, the predicted water stage of the next time, a Time Series method, and an Average Deviation method; utilizing a Kalman filter method to generate a random error of the water stage forecast; generating a water stage forecast correction of the next time according to the system error and the random error; and correcting a predicted water stage of the next time according to the water stage forecast correction and the predicted water stage.08-15-2013
20130212050AGENT APPARATUS FOR VEHICLE, AGENT SYSTEM, AGENT COLTROLLING METHOD, TERMINAL APPARATUS AND INFORMATION PROVIDING METHOD - An agent apparatus for a vehicle, an agent controlling method, a terminal apparatus and an information providing method, for executing a communication function with a personified agent, are provided. The apparatus includes an observing part that observes a driving situation based on sensor information; a learning part that learns by storing an observation result obtained from the observing part together with the sensor information; a determining part that determines a communication action with a user based on a learning result obtained from the learning part; a display control part that displays a first image in the vehicle expressing the communication action determined by the determining part; and an obtaining part that obtains acquired information acquired from the outside of the vehicle and stored in a portable terminal apparatus. The determining part may also determine the communication action by reflecting the acquired information on the learning result.08-15-2013
20130212047MULTI-TIERED APPROACH TO E-MAIL PRIORITIZATION - An apparatus for automating a prioritization of an incoming message, including a batch learning module that generates a global classifier based on training data that is input to the batch learning module. A feedback learning module that generates a user-specific classifier based on a plurality of feedback instances. A feature extraction module that receives the incoming message and a topic-based user model, infers a topic of the incoming message based on the topic-based user model, and computes a plurality of contextual features of the incoming message. A classification module that dynamically determines a priority classification strategy for assigning a priority level to the incoming message based on the plurality of contextual features of the incoming message and a weighted combination of the global classifier and the user-specific classifier, and classifies the incoming message based on the priority classification strategy.08-15-2013