Taxonomy of Clustering Methods Used in Fuzzy Logic Systems

Fuzzy logic systems have many applications in every field of moderate science. Most of the fuzzy logic systems are rule based reasoning, which are not easy to generate since the conflict between rules always arise in acquiring new knowledge. In recent years, there has been increasing interest in clustering-based fuzzy systems, which are easier to generate rules since they built from input-output training data. Clustering training data make the fuzzy system easier to maintain and more flexible in acquire real world knowledge. In this paper, we present taxonomy of clustering methods used in fuzzy logic systems. In particular, the exposition includes a discussion of strength and weakness of these methods and how they can be improved.