DEVELOPING A CASE-BASED KNOWLEDGE SYSTEM FOR AIDS-PREVENTION

Case-based reasoning (CBR) is a problem-solving paradigm where past ex periences are used to guide problem-solving. This paradigm shows a gre at deal of promise for use in intelligent systems. Recent work in inte lligent systems focusing on AIDS prevention reflects a growing interes t in the case-based paradigm because AIDS prevention experts rely heav ily on memory of previous cases when assessing subjects that exhibit A IDS-risky behaviors. If an AIDS prevention expert has seen a subject w ith similar AIDS-risky behavior previously, he or she is likely to dra w on that experience to propose a solution to the new case at hand. Th is paper describes a CBR system that functions as an AIDS prevention e xpert. The inputs are risk behavior descriptions and the subject's tes t results. The system employs fuzzy mathematical algorithms to retriev e and select previous cases, thereby assessing the subject's risk.