×

Method for refining cognitive insights using travel-related cognitive graph vectors

  • US 10,318,561 B2
  • Filed: 06/08/2015
  • Issued: 06/11/2019
  • Est. Priority Date: 06/09/2014
  • Status: Active Grant
First Claim
Patent Images

1. A cognitive inference and learning system for determining cognitive insights using cognitive graph vectors comprising:

  • a data store storing a cognitive graph including travel-related data from a plurality of data sources graph comprising a set of nodes and links between the nodes that represent travel-related knowledge;

    a processor; and

    a non-transitory computer readable medium including instructions for;

    associating a first subset of the set of nodes and links of the travel-related data within the cognitive graph with a first travel-related cognitive graph vector of a plurality of travel-related cognitive graph vectors, wherein the first travel-related cognitive graph vector comprises a plurality of first cognitive graph vector indices extending along the first travel-related cognitive graph vector away from a cognitive graph nexus, wherein a specificity of the first cognitive graph vector indices increases as they become more proximate to the cognitive graph nexus and each first cognitive graph vector index corresponding to a first travel-related attribute comprising one or more of who, how, why, where, when or what;

    associating a second subset of the set of nodes and links of the travel-related data within the cognitive graph with a second travel-related cognitive graph vector of the plurality of cognitive graph vectors, wherein the second travel-related cognitive graph vector comprises a plurality of second cognitive graph vector indices extending along the second travel-related cognitive graph vector away from the cognitive graph nexus, wherein a specificity of the second cognitive graph vector indices increases as they become more proximate to the cognitive graph nexus and each second cognitive graph vector index corresponding to a second travel-related attribute comprising one or more of who, how, why, where, when or what;

    receiving a user query;

    generating a first cognitive graph vector parameter for the first travel-related cognitive graph vector based on the user query;

    generating an initial travel-related cognitive insight from the cognitive graph based on the first cognitive graph vector parameter;

    defining a second cognitive graph vector parameter for the second travel-related cognitive vector based on the initial travel-related cognitive insight;

    generating a graph query based on the user query, the first cognitive graph vector parameter and the second cognitive graph vector parameter;

    querying a third subset of the set of nodes and links of the cognitive graph defined by the first cognitive graph vector and the second cognitive graph vector with the graph query to generate second travel-related cognitive insights; and

    returning the second travel-related cognitive insights in response to the user query.

View all claims
  • 8 Assignments
Timeline View
Assignment View
    ×
    ×