×

Topological methods to organize semantic network data flows for conversational applications

  • US 6,778,970 B2
  • Filed: 05/28/1998
  • Issued: 08/17/2004
  • Est. Priority Date: 05/28/1998
  • Status: Expired due to Term
First Claim
Patent Images

1. A method executed by a data processor system for quantifying clarity in a natural language processing system by measuring abstractness deviation for sets of inheritance sibling nodes within a semantic inheritance network, comprising the steps of:

  • a) identifying a set of inheritance sibling nodes;

    b) computing an average abstractness for the set of inheritance sibling nodes;

    c) computing an abstractness deviation from the average abstractness, for each inheritance sibling node in the set of inheritance sibling nodes;

    d) summing the abstractness deviations of each inheritance sibling node in the set of inheritance sibling nodes;

    e) comparing the summed abstractness deviation of a set of sibling nodes to a summed abstractness deviation of an alternative topology for the set of inheritance nodes; and

    f) optimizing the semantic network inheritance link topology by selecting the alternative topology with less abstractness deviation.

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