A New Recognition System of Textual Entailment Exploiting Wide-Coverage Lexical Knowledge

Shane Mccrae

Abstract


In this article, we aim to present a predicate-argument Textual Entailment Recognition system which ia able to exploit wide-coverage lexical knowledge. in contrast to conventional machine learning approaches where several features obtained from linguistic analysis and resources are utilized, method introduced in this paper regards a predicate-argument structure as a basic unit. Additionally, it performs the matching/alignment between a text and hypothesis, which are automatically acquired from a dictionary, Web corpus, and Wikipedia.

Refbacks

  • There are currently no refbacks.


Paper submission email: AJLL@KnowledgeAustralia.com

ISSN (Paper)application pending

ISSN (Online)application pending