Case-based reasoning systems: from automation to decision-aiding and stimulation

Over the past decade, case-based reasoning (CBR) has emerged as a major research area within the artificial intelligence research field due to both its widespread usage by humans and its appeal as a methodology for building intelligent systems. Conventional CBR systems have been largely designed as automated problem-solvers for producing a solution to a given problem by adapting the solution to a similar, previously solved problem. Such systems have had limited success in real-world applications. More recently, there has been a search for new paradigms and directions for increasing the utility of CBR systems for decision support. The paper focuses on the synergism between the research areas of CBR and decision support systems (DSSs). A conceptual framework for DSSs is presented and used to develop a taxonomy of three different types of CBR systems: 1) conventional, 2) decision-aiding, and 3) stimulative. The major characteristics of each type of CBR system are explained with a particular focus on decision-aiding and stimulative CBR systems. The research implications of the evolution in the design of CBR systems from automation toward decision-aiding and stimulation are also explored.