A rule discovery by fuzzy classifier system utilizing symbolic information

摘要

This paper presents a new fuzzy classifier system (FCS) that can discover effective fuzzy rules efficiently. The system incorporates human knowledge in the form of symbolic information, and effectively limits its search space for fuzzy rules by using knowledge. The system also extracts symbolic information from acquired fuzzy rules for efficient exploration of other new fuzzy rules. Simulations are done to demonstrate the feasibility of the proposed method.