Using the web in machine learning for other-anaphora resolution

  • Authors:
  • Natalia N. Modjeska;Katja Markert;Malvina Nissim

  • Affiliations:
  • University of Edinburgh;University of Leeds;University of Edinburgh

  • Venue:
  • EMNLP '03 Proceedings of the 2003 conference on Empirical methods in natural language processing
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present a machine learning framework for resolving other-anaphora. Besides morpho-syntactic, recency, and semantic features based on existing lexical knowledge resources, our algorithm obtains additional semantic knowledge from the Web. We search the Web via lexico-syntactic patterns that are specific to other-anaphors. Incorporating this innovative feature leads to an 11.4 percentage point improvement in the classifier's F-measure (25% improvement relative to results without this feature).