Similarity measures for retrieval in case-based reasoning systems

Abstract

Case-based reasoning (CBR) is one of the emerging paradigms for designing intelligent systems. Retrieval of similar cases is a primary step in CBR, and the similarity measure plays a very important role in case retrieval. Sometimes CBR systems are called similarity searching systems, the most important characteristic of which is the effectiveness of the similarity measure used to quantify the degree of resemblance between a pair of cases. This article focuses on the similarity measuring methods for CBR and comprises two parts. The first part reviews the existing methods for measuring similarity in the literature based on more than 100 CBR project studies and some general similarity measures seen in other applications. In the second part, a hybrid similarity measure is proposed for comparing cases with a mixture of crisp and fuzzy features. Its application to the domain of failure analysis is illustrated.