Patent application number | Description | Published |
20080229828 | Establishing reputation factors for publishing entities - The architecture utilizes the network effects of patents, journals, authors, institutions, and funding entities, for example, to establish an objective reputation factor. The reputation factor contributes to a higher perceived relevance as well as provides interesting new services that could be built on top. The algorithm takes into account not only the number of cited-by references for a certain paper, author, or institution, but can also generate a higher ranking for cross-disciplinary citations, citations establishing a new area of science, acknowledgement citations, and constantly-updated reputation factors of different important entities, such as co-authorship, institutional affiliation, and journal impact factor. Impact factors can be fed back into the system for consideration in generating the reputation factor. | 09-25-2008 |
20080319749 | GENERIC SPELLING MNEMONICS - A system and method for creating a mnemonics Language Model for use with a speech recognition software application, wherein the method includes generating an n-gram Language Model containing a predefined large body of characters, wherein the n-gram Language Model includes at least one character from the predefined large body of characters, constructing a new Language Model (LM) token for each of the at least one character, extracting pronunciations for each of the at least one character responsive to a predefined pronunciation dictionary to obtain a character pronunciation representation, creating at least one alternative pronunciation for each of the at least one character responsive to the character pronunciation representation to create an alternative pronunciation dictionary and compiling the n-gram Language Model for use with the speech recognition software application, wherein compiling the Language Model is responsive to the new Language Model token and the alternative pronunciation dictionary. | 12-25-2008 |
20090106232 | BOOSTING A RANKER FOR IMPROVED RANKING ACCURACY - A system described herein includes a trainer component that receives an estimated gradient of cost that corresponds to a first ranker component with respect to at least one training point and at least one query. The trainer component builds a second ranker component based at least in part upon the received estimated gradient. The system further includes a combiner component that linearly combines the first ranker component and the second ranker component. | 04-23-2009 |
20090125701 | AGGREGATING DATA FROM DIFFERENT SOURCES - A method and system that aggregates data associated with one or more entities from different data sources are provided. The data sources include documents, web pages, or images that have information about one or more entities. The information is extracted from the data sources based on criteria that define the entities. The extracted information is utilized to generate a hash identifier that corresponds to each entity and one or more storage locations. The one or more storage locations and associated hash identifiers are utilized to store the extracted information corresponding to the entities, and the extracted information for each entity is structured as a virtual page that is stored in an index having references to the data sources. The index storing the virtual pages is notified or updated when the associated data sources are modified. | 05-14-2009 |
20100262612 | RE-RANKING TOP SEARCH RESULTS - The claimed subject matter provides a system and/or a method that facilitates generating sorted search results for a query. An interface component can receive a query in a first language. A first ranker can be trained from a portion of data related to a second language. A second ranker can correspond to the first language, wherein the second ranker is untrained due to a limited amount of data related to the first language. A sorting component can invoke the first ranker to generate and order a pre-defined number of search results for the received query and subsequently invoke the second ranker to the pre-defined number of search results to generate a re-ordered number of search results in the first language for the received query. | 10-14-2010 |
20100299350 | CLICK-THROUGH PREDICTION FOR NEWS QUERIES - Described is estimating whether an online search query is a news-related query, and if so, outputting news-related results in association with other search results returned in response to the query. The query is processed into features, including by accessing corpora that corresponds to relatively current events, e.g., recently crawled from news and blog articles. A corpus of static reference data, such as an online encyclopedia, may be used to help determine whether the query is less likely to be about current events. Features include frequency-related data and context-related data corresponding to frequency and context information maintained in the corpora. Additional features may be obtained by processing text of the query itself, e.g., “query-only” features. | 11-25-2010 |
20120323876 | SEARCH RESULTS BASED ON USER AND RESULT PROFILES - A method includes receiving a search query from a user at a search engine. The method also includes performing a search responsive to the search query to obtain a set of results. The set of results is ordered at least partially based on a distance between a profile of a user and a profile of each result in the ordered set of results. At least one profile of a result is based on an aggregate of a plurality of user profiles. | 12-20-2012 |
20130268513 | ANNOTATIONS BASED ON HIERARCHICAL CATEGORIES AND GROUPS - Systems and methods for recommending entities to a user are presented. In at least one embodiment, a user is identified as belonging to one or more groups of users, or to a hierarchy of groups. A category of entities, preferred by at least of the hierarchy of identified groups, is identified. The category of entities preferred by the at least one identified group corresponds to the category of a user-preferred entity preferred by the user. An entity from the category of entities is selected; the selected entity is not the user-preferred entity. The selected entity is provided to the user as a recommended entity to the user. | 10-10-2013 |
20140129505 | SOCIAL EVENT RECOMMENDATION SYSTEM - A system for recommending social events includes a front-end interface for establishing communication over a communications network and a user database module for accessing a plurality of user profiles. An activities database module is provided for accessing a plurality of activities that users may attend. The system also includes a recommendation engine operable in at least a first user-selectable mode. The first user-selectable mode provides at least one recommendation. Each of the recommendations includes a plurality of invitees and at least one activity to be attended by the plurality of invitees. The invitees are selected from a set of invitees accessible through the user database module and the activity is selected from a set of activities accessible through the activities database module. In one mode of operation, the invitees and the activity are both selected by the system and are not preselected by the user. | 05-08-2014 |
20140188866 | RECOMMENDATION ENGINE BASED ON CONDITIONED PROFILES - A conditional recommendation system and methods are presented. The conditional recommendation engine (embodied as a system, a method, or computer-readable media) provides recommendations of entities to a computer user based on a conditioned user profile for the computer user. The conditioned user profile identifies user preferences in regard to dynamic, changeable conditional factors. These factors may include the time of day that a request was made, the day of week, the month of the year, the season, whether it is a holiday and which holiday, and the like. | 07-03-2014 |