Cooperation between Pronoun and Reference Resolution for Unrestricted Texts

TL;DR

This paper presents a flexible architecture for anaphora resolution that leverages character cues and can emulate classic methods, with promising results on narrative texts.

Contribution

It introduces a configurable open architecture for anaphora resolution that integrates character cues and can replicate existing methods.

Findings

Narrative texts are most suitable for testing the system.

The system effectively uses minimal resources like lexicons and parsers.

The approach improves pronoun resolution accuracy.

Abstract

Anaphora resolution is envisaged in this paper as part of the reference resolution process. A general open architecture is proposed, which can be particularized and configured in order to simulate some classic anaphora resolution methods. With the aim of improving pronoun resolution, the system takes advantage of elementary cues about characters of the text, which are represented through a particular data structure. In its most robust configuration, the system uses only a general lexicon, a local morpho-syntactic parser and a dictionary of synonyms. A short comparative corpus analysis shows that narrative texts are the most suitable for testing such a system.