Social network model for semantic processing

摘要

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.

主权项

1. A computer-implemented method of generating a user's social network model, the method comprising:
receiving a set of documents associated with a user; accessing the user's contact data, the contact data identifying a plurality of entities; analyzing the documents, using the contact data, to identify references to entities therein; identifying relationships among the referenced entities; determining a strength of a first relationship between a first entity and a second entity responsive to a volume of documents in which both the first entity and the second entity appear, wherein the first and second entities are a subset of the referenced entities; building a social network model for the user responsive to the identified relationships among the referenced entities and the strength of the first relationship; storing the social network model; receiving a new document associated with the user; identifying, in the new document, a reference to an ambiguous entity; performing name disambiguation using the social network model to determine which of at least two candidate entities from the social network model is an intended entity for the ambiguous entity; identifying other entities referenced by the new document; and updating the social network model by modifying relationship strengths in the social network model between the intended entity and the other entities referenced by the new document.