Understanding new metaphors

  • Authors:
  • James H. Martin

  • Affiliations:
  • Berkeley Artificial Intelligence Research, Computer Science Division, University of California, Berkeley, California

  • Venue:
  • IJCAI'87 Proceedings of the 10th international joint conference on Artificial intelligence - Volume 1
  • Year:
  • 1987

Quantified Score

Hi-index 0.00

Visualization

Abstract

Metaphors are widespread in natural language. Most natural language processing systems have failed to adequately deal with this issue. A two part approach to this problem involving a Metaphoric Lexicon and a dynamic extension system is described. The extension mechanism operates by analogically mapping previously understood mappings to account for new uses based on known regularities in the English metaphor system. A system based on this approach has been applied to the UNIX domain. Results indicate that this domain exhibits all the predicted regularities and that understanding new senses is possible using these regularities