×

Method and system for implementing semantic analysis of internal social network content

  • US 9,727,925 B2
  • Filed: 09/09/2013
  • Issued: 08/08/2017
  • Est. Priority Date: 09/09/2012
  • Status: Active Grant
First Claim
Patent Images

1. A method implemented with a processor, comprising:

  • identifying an internal social network for an enterprise;

    collecting a set of messages from the internal social network;

    performing semantic filtering to the set of messages, the semantic filtering reducing an excess noise associated with data that is not relevant to the enterprise;

    performing semantic analysis upon the set of messages collected from the internal social network to determine a contextual significance of one or more terms in the set of messages;

    identifying themes in the set of messages, the themes pertaining to at least one of customer preferences, demographic information, industry trends, customer view points;

    as a result of the performed semantic analysis, clustering together messages that are similar to each other;

    categorizing the set of messages into a plurality of categories based at least in part on the contextual significance of the one or more terms in the set of messages;

    associating each category of the plurality of categories with one or more tags;

    associating respective tags to the set of messages that are categorized as a result of the semantic analysis to generate a set of tagged messages;

    determining associations between the one or more tags and one or more enterprise applications corresponding to the enterprise such that the set of tagged messages are sent to respective ones of the one or more enterprise applications based at least in part on the determined associations; and

    storing the tagged messages in an actionable social message store, the actionable message store storing in a table format, for each message, one or more of a source of the message, a topic parameter associated with the message, data associated with the message, and the one or more tags associated with the message, wherein the semantic analysis performed comprises latent semantic analysis (LSA), the LSA referring to a form of statistical language modeling that distinguishes two identical words based on a semantic significance of the word.

View all claims
  • 1 Assignment
Timeline View
Assignment View
    ×
    ×