Coreference resolution with syntactico-semantic rules and corpus statistics

  • Authors:
  • Sven Hartrumpf

  • Affiliations:
  • University of Hagen, Hagen, Germany

  • Venue:
  • ConLL '01 Proceedings of the 2001 workshop on Computational Natural Language Learning - Volume 7
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

A new hybrid approach to the coreference resolution problem is presented. The CORUDIS system (COreference RUles with DIsambiguation Statistics) combines syntactico-semantic rules with statistics derived from an annotated corpus. First, the rules and corpus annotations are described and exemplified. Then, the coreference resolution algorithm and the involved statistics are explained. Finally, the proposed method is evaluated against a baseline model and some directions for further research are indicated.