×

Measuring semantic incongruity within text data

  • US 10,176,260 B2
  • Filed: 02/10/2015
  • Issued: 01/08/2019
  • Est. Priority Date: 02/12/2014
  • Status: Active Grant
First Claim
Patent Images

1. A method comprising:

  • processing, by a computing device, textual data associated with a digital object to identify within the textual data one or more semantic exemplars for the digital object;

    generating, by the computing device, a data structure that represents a semantic space of semantic diversity and semantic anomalousness for the digital object and specifies a respective measure of similarity between each pairwise combination of the semantic exemplars identified within the textual data;

    computing, by the computing device and based at least in part on the data structure, a level of incongruity for the digital object that represents a predicted estimation of interest of the digital object to a user and from a function that quantifies joint presence of semantic diversity and semantic anomalousness; and

    executing a programmatic action based on the computed level of incongruity for the digital object, wherein executing the programming action comprises outputting, by the computing device, an indication of the level of incongruity for the digital object.

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