Patent application number | Description | Published |
20130031034 | ADAPTIVE 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 |
20130031489 | NEWS FEED RANKING MODEL BASED ON SOCIAL INFORMATION OF VIEWER - 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 |
20140258191 | ADAPTIVE 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. | 09-11-2014 |
Patent application number | Description | Published |
20130290430 | AGGREGATING SOCIAL NETWORKING SYSTEM USER INFORMATION FOR DISPLAY VIA STORIES - To generate stories for presentation to social networking system users, one or more story generation engines receives a set of actions of social networking system users. Various story generation engines output stories of different types that collect particular types of actions and arrange the actions in a specified format for presentation. The story generators may avoid selecting the same actions for different stories. Stories generated by the various story generators may be ranked and the ranking used to select stories for presentation to one or more social networking system users. | 10-31-2013 |
20140019261 | SPONSORED ADVERTISEMENT RANKING AND PRICING IN A SOCIAL NETWORKING SYSTEM - A social networking system (SNS) provides sponsored stories and organic stories about actions taken by other SNS users to a viewing user. Organic stories are selected based on the likelihood the viewing user is interested in their content. While advertisers compensate the SNS for presentation of sponsored stories, the sponsored stories also include information about actions by other SNS users. To increase the likelihood the viewing user interacts with sponsored stories, a common communication channel is used to present both the sponsored stories and the organic stories. To simplify selection of organic stories and sponsored stories, the SNS determines a common unit of measurement for both and makes selections based on the common unit of measurement. | 01-16-2014 |
20140032564 | Social Static Ranking for Search - In one embodiment, one or more computing devices assign each of a plurality of nodes of a graph of a social-networking system to one of a plurality of search indices. Each search index corresponds to a node type, and each node assigned to a search index is of the node type that the search index corresponds to. For each search index, the one or more computing devices determine a value for each node assigned to the search index based at least in part on edges connected to the node in the graph and rank the nodes assigned to the search index based at least in part on their values. The one or more computing devices provide the search indices for storage to facilitate responding to queries encompassing objects represented by the nodes assigned to the search indices. | 01-30-2014 |
20150088872 | Social Static Ranking for Search - In one embodiment, one or more computing devices receive, from a client device of a first user, a query from the first user. The computer devices search a social graph to identify one or more nodes of the social graph that are relevant to the query. The computer devices obtain a static rank for each identified node. The static rank is based at least in part on a number of edges of a particular edge type that are connected to the node in the graph or attributes of edges connected to the node in the graph. The computer devices send to the client device of the first user for display, a search-results page responsive to the received query. The search-results page includes reference to one or more nodes having a static rank greater than a threshold rank. | 03-26-2015 |
20150324339 | PROVIDING FACTUAL SUGGESTIONS WITHIN A DOCUMENT - Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing suggestions within a document. In one aspect, a method includes obtaining textual input provided to a document editing application by a user device, the textual input being provided to the document editing application for inclusion in a document, and wherein the document includes prior text that was included in the document prior to the textual input; identifying an entity based on entity text included in the textual input; identifying an attribute of the entity based on attribute text included in the textual input; generating a query specifying the entity and the attribute; providing the query to a search system that provides a result value for the attribute of the entity included in the query; and providing the result value to the user device as a suggestion for inclusion in the document. | 11-12-2015 |
20150324413 | UPDATING TEXT WITHIN A DOCUMENT - Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for updating text within a document. In one aspect, a method includes identifying an entity based on entity text included in document text of a document; identifying a time-sensitive attribute for the entity based on attribute text included in the document text; identifying a first value for the time-sensitive attribute based on value text included in the document text; generating a query specifying the entity; providing the query to a search system that provides a result value for the time-sensitive attribute of the entity included in the query; and providing, to a user device that is currently accessing the document, result data that causes presentation of the result value as a replacement for the first value. | 11-12-2015 |