Case Base Management Systems

Abstract

Case-based reasoning offers a new approach for developing knowledge-based systems. Unlike the rule-based paradigm, in which domain knowledge is encoded in the form of production rules, in the case-based approach the problem solving experience of the domain expert is captured by cases. These cases are stored in a knowledge base, called a casebase. Current case-based systems are mostly research prototypes (programming-in-the-small), designed primarily to refine the concepts in case-based reasoning. A number of research issues need to be resolved to facilitate the transition of these prototypes to large application systems (programming-inthe-large). The primary issue that will help the transition is to develop data management support for these prototypes. In this paper, we analyze this data management support and propose a new concept called a casebase management system to perform data management functions for case-based systems. We discuss the research issues that should be addressed to develop casebase management systems. Case-based reasoning offers a novel approach to develop knowledge-based systems. Unlike the rule-based paradigm in which domain knowledge is encoded in the form of rules, in the case-based approach the expert's problem solving experience is captured in the form of cases which are stored in a casebase. Case-based systems solve problems by reusing problem solving experience stored in cases. These systems are especially suitable for domains with the following characteristics [Kolodner & Riesbeck, 1990] : • problems are ill defined and/or incompletely specified; • knowledge used for problem solving is obscure; • it is difficult to formulate domain rules, but examples are