Topic Detection Using Lexical Chains

  • Authors:
  • Yllias Chali

  • Affiliations:
  • -

  • Venue:
  • Proceedings of the 14th International conference on Industrial and engineering applications of artificial intelligence and expert systems: engineering of intelligent systems
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper describes an algorithm for identifying the topic of unrestricted texts. The algorithm takes as input segments of text that represent grouping of contiguous portions of the text, and discovers lexical chains as indicator of their topics. Two implementation, based on public domain resources, are presented: one based on WordNet and the second one based on Roget's thesaurus. The evaluation of the algorithm shows that lexical chains are acceptable as topic indicator with 45% of precision and 65% of recall.