A probabilistic genre-independent model of pronominalization

  • Authors:
  • Michael Strube;Maria Wolters

  • Affiliations:
  • European Media Laboratory GmbH, Heidelberg, Germany;Universität Bonn, Bonn, Germany

  • Venue:
  • NAACL 2000 Proceedings of the 1st North American chapter of the Association for Computational Linguistics conference
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

Our aim in this paper is to identify genreindependent factors that influence the decision to pronominalize. Results based on the annotation of twelve texts from four genres show that only a few factors have a strong influence on pronominalization across genres, i.e. distance from last mention, agreement, and form of the antecedent. Finally, we describe a probabilistic model of pronominalization derived from our data.