FORMALIZING PARTIAL MATCHING AND SIMILARITY IN CASE-BASED REASONING WITH A DESCRIPTION LOGIC

Our aim is to use a description logic including default delta and exce ption epsilon connectives as a formal framework for a case-based reaso ning (CBR) system. This approach allows the retrieval of similar cases to be formalized Subsumption and (sure, probable, typical, and except ional) inheritance relations of the description logic are the foundati ons for the different retrieval tasks: abstracting the new case; class ifying it in the index base (full and partial matching); evaluating si milarity of the conceptual abstraction of the new case with the concep ts of the index base, using conceptual preference criteria; and retrie ving similar cases (instances) and applying instance preference criter ia to order them. Our preference criteria are symbolic rather than num erical or those of fuzzy logic. Using description logic offers several advantages: the classification process call be used to retrieve simil ar cases, the formal properties and the efficiency of the system can e asily be evaluated, and preference criteria are homogeneously based on the formal description logic framework. Moreover preference criteria are independent of the knowledge and can thus be used in other applica tions.