A KNOWLEDGE BASE SYSTEM FOR DECISION SUPPORT USING COGNITIVE MAPS

Abstract This paper presents an man-computer system for decision support using cognitive maps. we first summarize the method of documentary coding and derivation of cognitive maps. Second, we construct a knowledge base by joining cognitive maps of five documents relevant to the traffic problem in Japan and indicate that some kinds of useful information are generated in the knowledge base through the process of joining. Third, we propose three retrieval modes from knowledge base. Those retrieval modes provide the following knowledge according to types of the decisionmakers' requests: 1) Skeleton maps indicate overall causal structure of the problem; 2) Hierarchical graphs give detailed information about parts of the causal structure; 3) Sources of causal relations are presented when necessary, for example when the decisionmaker wants to browse the causal assertions in documents. Using them, we describe an illustrative example of how the knowledge base system assists the decisionmaker in grasping a complex problem.