F4.3 Fuzzy Case-Based Reasoning Systems

Case-based Reasoning (CBR), a method of analogical reasoning which is common and extremely important in human cognition, has only recently emerged as a major reasoning methodology. CBR involves solving new problems by identifying and adapting solutions to similar problems stored in a library of past experiences. The important steps in the inference cycle of CBR are to retrieve cases from the library which are most relevant to the problem at hand and adapt the retrieved cases to the current input. First we will review the history of CBR and discuss two types of CBR systems (interpretive and problem-solving). Then, we will analyze the most important CBR issues, such as case retrieval and selection, memory organization, matching and similarity measures, case adaptation, evaluation, and integration. Finally, we will explore some of Fuzzy Logic contributions to CBR and focus on the computation of abstract features for case indexing and their use in computing similarity measures to retrieve the most relevant cases.