Résumé

Ce papier propose une approche de recherche d’information basée sur l’utilisation

Abstract

The paper proposes an approach to information retrieval based on the use of a conceptual structure both for indexing document and expressing user queries. The conceptual structure is hierarchical and it is formally represented as a weighted tree. In this approach, the evaluation of queries is based on the comparison of minimal sub-trees containing the two sets of nodes corresponding to the concepts expressed in the document and the query respectively. The comparison is based on the computation of a multiple-valued degree of inclusion. Some candidate implications are discussed on the basis of their respective semantics. The proposed approach generalizes standard fuzzy information retrieval and its evaluation on benchmark example is also presented.