Skip to Main content Skip to Navigation

An Integrated Computational Approach to Binding Theory

Roberto Bonato 1, 2, 3
3 SIGNES - Linguistic signs, grammar and meaning: computational logic for natural language
INRIA Futurs, Université Sciences et Technologies - Bordeaux 1, École Nationale Supérieure d'Électronique, Informatique et Radiocommunications de Bordeaux (ENSEIRB), Université Bordeaux Montaigne, CNRS - Centre National de la Recherche Scientifique : UMR5800
Abstract : Pronouns play a decisive role in every natural language as the linguistic ele- ments that enable semantic cohesion of a text. Anaphora resolution (that is, the task of recovering in an automatic way the semantic content of pronouns) is therefore both an important theoretical issue and a ma jor technological challenge for any computer application that aims at a ?ner-grained semantic analysis of natural language texts. Binding Theory is a branch of generative linguistics devoted to discovering the principles that rule the distribution and interpretation of pronouns in a sentence. In this thesis, we present two algorithms for anaphora resolution inspired by three of the most influential approaches to Binding Theory in linguistics. Eventually, we combine insights issued from last 30 years of linguistic enquiry into an original computational framework and integrate them into a new algorithm. We also compare our computational approach with the purely semantic one recently proposed by Philippe Schenker.
Complete list of metadatas

Cited literature [38 references]  Display  Hide  Download
Contributor : Christian Retoré <>
Submitted on : Sunday, September 20, 2009 - 3:42:25 PM
Last modification on : Tuesday, July 9, 2019 - 10:12:55 AM
Long-term archiving on: : Tuesday, June 15, 2010 - 11:55:46 PM


  • HAL Id : tel-00418563, version 1


Roberto Bonato. An Integrated Computational Approach to Binding Theory. Computer Science [cs]. Université Sciences et Technologies - Bordeaux I; Università degli studi di Verona, 2006. English. ⟨tel-00418563⟩



Record views


Files downloads