Raisonnement à base de cas textuels Etat de l'art et perspectives

Traditionnally case-based reasoning is conducted on experiences that are represented in a well structured format such as objects or database records. However different models have been proposed to overcome some of the limitations imposed by this structural approach and to undertake new applications domains. In this paper, we review some of the extensions proposed to apply CBR principles to experiences contained in textual documents, commonly refered to as textual CBR. Following a short presentation of some case-based reasoning principles, we present the main research works in Textual CBR and provide a synthesis of current state of the art in the field. We finally discuss some shortcomings of the current approaches and propose some directions for future research. MOTS-CLÉS : raisonnement à base de cas, cas textuels, adaptation, recherche d’information.