GIR experimentation

  • Authors:
  • Andogah Geoffrey

  • Affiliations:
  • Computational Linguistics Group, Centre for Language and Cognition Groningen, University of Groningen, Groningen, The Netherlands

  • Venue:
  • CLEF'06 Proceedings of the 7th international conference on Cross-Language Evaluation Forum: evaluation of multilingual and multi-modal information retrieval
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

The Geographic Information Retrieval (GIR) community has generally accepted the thesis that both thematic and geographic aspects of documents are useful for GIR. This paper describes an experiment exploring this thesis by separately indexing and searching geographical relevant-terms (place names, geo-spatial relations, geographic concepts and geographic adjectives). Two indexes were created - one for extracted geographic relevant-terms (footprint document) and one for reference document collections. Footprint and reference document scores are combined by a linear interpolation to obtain an overall score for document relevance ranking. Experimentation with geographic query expansion provided no significant improvement though it stabilized the query result.