Combination of case-based reasoning and data mining through integration with the domain ontology

Abstract

In present paper we consider a combination of CBR with the Domain Ontology and Data Mining. The proposed approach to integration CBR with the Ontology, as a result, gives a numerical matrix that characterize the semantic relationship of cases with the domain concepts. Then it is possible to apply methods of cluster analysis, classification and regression to this data matrix in order to organize semantic space of cases and thus increase the relevance of their retrieval. We explore possibilities of fuzzy logic approaches for building fuzzy rules and classification trees.