Constructing Fuzzy Models from Partitions

As shown in the previous chapter, a class of fuzzy clustering algorithms can be used to approximate a set of data by local linear models. Each of these models is represented by a fuzzy subset in the data set available for identification. In order to obtain a model useful for prediction or controller design, an additional step must be applied to generate a model independent of the identification data. Such a model can be represented either as a rule base or as a fuzzy relation. This chapter presents methods and algorithms for constructing fuzzy rule-based and relational models from the fuzzy partitions obtained by product-space clustering.