LEXPIR: A VERB LEXICON

Authors

  • Monserrat Civit
  • Irene Castellón
  • Mª Antonia Martí
  • Mariona Taulé

Keywords:

Information extraction, semantic domain, semantic representation, verb lexicon

Abstract

The present work lies within the area of information Extraction (IE) Systems deal with restricted semantic domains and have been mostly developed for English. This paper describes LEXPIR, a Spanish verb lexicon that plays a central role in an IE system that is based on linguistic knowledge, and is capable of dealing with unrestricted domains. This verb lexicon has been developed following the theoretically founded model of verb lexical entry, from which to derive the predicate semantic classification. In additiori to LEXPIR, the IE system architecture includes a syntactic analyser (TACAT), a morphological analyser (MACO) and a semantic riel (EuroWordNet). The system helps to obtain a semantic representation for the text basic contents

Author Biographies

Monserrat Civit

Centre de Recerca TALP Universitat Politécnica de Cataluya

Irene Castellón

Departament de Lingüística General Universitat de Barcelona

Mª Antonia Martí

Departament de Lingüística General Universitat de Barcelona

Mariona Taulé

Departament de Lingüística General Universitat de Barcelona

Issue

Section

Artículos