Language Resources in Cross-Language Text Retrieval: A CLEF Perspective

  • Authors:
  • Julio Gonzalo

  • Affiliations:
  • -

  • Venue:
  • CLEF '00 Revised Papers from the Workshop of Cross-Language Evaluation Forum on Cross-Language Information Retrieval and Evaluation
  • Year:
  • 2000

Quantified Score

Hi-index 0.01

Visualization

Abstract

Language resources such as machine dictionaries and lexical databases, aligned parallel corpora or even complete machine translation systems are essential in Cross-Language Text Retrieval (CLTR), although not standard tools for the Information Retrieval task in general. We outline the current use and adequacy for CLTR of such resources, focusing on the participants and experiments performed in the CLEF 2000 evaluation. Our discussion is based on a survey conducted on the CLEF participants, as well as the descriptions of their systems that can be found in the present volume. We also discuss how the usefulness of the CLEF evaluation campaign could be enhanced by including additional tasks which would make it possible to distinguish between the effect on the results of the resources used by the participating systems, on the one hand, and the retrieval strategies employed, on the other.