Supervised fuzzy clustering for rule extraction

This paper deals with the application of orthogonal transforms and fuzzy clustering to extract fuzzy rules from data. It is proposed to use the orthogonal least squares method to supervise the progress of the fuzzy clustering algorithm and remove clusters of less importance with respect to describing the data. Clustering takes place in the product space of systems inputs and outputs and each cluster corresponds to a fuzzy IF-THEN rule. By initializing the clustering with an overestimated number of clusters and subsequently remove less important ones as the clustering progresses, it is sought to obtain a suitable partition of the data in an automated manner. The approach is generally applicable to the fuzzy c-means and related algorithms. The adaptive distance norm fuzzy clustering is studied and applied to the identification of Takagi-Sugeno type rules. Both a synthetic example as well as a real-world modeling problem are considered to illustrate the working and the applicability of the algorithm