×

Determining query return referents for concept types in conceptual graphs

  • US 9,158,838 B2
  • Filed: 12/15/2008
  • Issued: 10/13/2015
  • Est. Priority Date: 12/15/2008
  • Status: Active Grant
First Claim
Patent Images

1. A method, implemented in a computer, for determining query return referent data for concept types in a conceptual graph, the computer comprising a processor and a first memory configured to store a plurality of instructions executable by the processor to implement the method, wherein the method comprises:

  • generating a conceptual graph for a document query, the conceptual graph comprising a plurality of graph terms;

    identifying, in the conceptual graph, at least a first respective graph term having unknown referent data, the unknown referent data comprising an unknown specific instance of a respective concept type associated with the first respective graph term having unknown referent data;

    identifying, for the first respective graph term having unknown referent data, at least a second respective graph term that corresponds to a linking concept term in the conceptual graph, wherein the identification of the linking concept term is based at least in part on determining a function of the second respective graph term in the conceptual graph, wherein the second respective graph term is identified as the linking concept term if the second respective graph term functions to provide a context and relationship between the first respective graph term having unknown referent data and at least a third graph term;

    accessing a first knowledge base comprising a plurality of mappings of conceptually similar terms to the graph terms in the conceptual graph;

    identifying, based at least in part on information in the plurality of mappings, conceptually similar terms for each of the graph terms in the conceptual graph;

    for at least the first respective graph term having unknown referent data, modifying a corresponding mapping to the first respective graph term having unknown referent data in the first knowledge base, to identify the first respective graph term having unknown referent data as being a first respective graph term having unknown referent data that is in need of at least respective potential referent data;

    for at least the second respective graph term, modifying a corresponding mapping to the second respective graph term in the first knowledge base, to identify the second respective graph term as the linking concept term;

    for at least the first respective graph term having unknown referent data, identifying respective potential referent data to be added to the conceptual graph by searching the first knowledge base, using the modified mappings to the first and second graph terms, respectively, for instances where conceptually similar terms to the first respective graph term having unknown referent data are associated, in at least one other mapping, with conceptually similar terms for the linking concept term, wherein respective potential referent data that has been identified is referred to as respective identified potential referent data;

    associating, in the modified mapping for the first respective graph term having unknown referent data, the respective identified potential referent data with the first respective graph term having unknown referent data; and

    validating the conceptual graph based at least in part on analyzing a validity of each respective identified potential referent data.

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