Towards a Possibilistic Information Retrieval System Using Semantic Query Expansion

  • Authors:
  • Bilel Elayeb;Ibrahim Bounhas;Oussama Ben Khiroun;Fabrice Evrard;Narjès Bellamine-BenSaoud

  • Affiliations:
  • ENSI Manouba University, Tunisia;Faculty of Sciences of Tunis, Tunisia;ENSI Manouba University, Tunisia;Informatics Research Institute of Toulouse IRIT, France;ENSI Manouba University, Tunisia

  • Venue:
  • International Journal of Intelligent Information Technologies
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents a new possibilistic information retrieval system using semantic query expansion. The work is involved in query expansion strategies based on external linguistic resources. In this case, the authors exploited the French dictionary "Le Grand Robert". First, they model the dictionary as a graph and compute similarities between query terms by exploiting the circuits in the graph. Second, the possibility theory is used by taking advantage of a double relevance measure possibility and necessity between the articles of the dictionary and query terms. Third, these two approaches are combined by using two different aggregation methods. The authors also benefit from an existing approach for reweighting query terms in the possibilistic matching model to improve the expansion process. In order to assess and compare the approaches, the authors performed experiments on the standard 'LeMonde94' test collection.