Selecting Fuzzy Rules with Forgetting in Fuzzy Classification Systems

This paper proposes a rule selection method with the destructive learning algorithm to construct a compact fuzzy classification system with high performance. In this paper, first we construct a fuzzy classification system by generating fuzzy rules from numerical data, and consider the fuzzy classification system based on fuzzy rules a network. Then we select significant fuzzy rules from the rule set by the proposed method which can remove unnecessary fuzzy rules. We demonstrate the effectiveness of the proposed method by applying it to the classification problem of the iris data of Fisher. Figure 1: Simple fuzzy grid