Composition of semantic relations: model and applications

  • Authors:
  • Eduardo Blanco;Hakki C. Cankaya;Dan Moldovan

  • Affiliations:
  • The University of Texas at Dallas;The University of Texas at Dallas;The University of Texas at Dallas

  • Venue:
  • COLING '10 Proceedings of the 23rd International Conference on Computational Linguistics: Posters
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents a framework for combining semantic relations extracted from text to reveal even more semantics that otherwise would be missed. A set of 26 relations is introduced, with their arguments defined on an ontology of sorts. A semantic parser is used to extract these relations from noun phrases and verb argument structures. The method was successfully used in two applications: rapid customization of semantic relations to arbitrary domains and recognizing entailments.