Automated Methods for Fuzzy Systems
Timothy J. Ross · 2010
Fuzzy modeling is very practical and can be used to develop a model for the system using the ‘limited' available information. Batch least squares (BLS), recursive least squares (RLS), gradient method (GM), learning from example (LFE), modified learning from example (MLFE), and clustering method (CM) are some of the algorithms available for developing a fuzzy model. These methods, which are referred to as automated methods, are provided as additional procedures to develop membership functions. This chapter summarizes these six methods for use in developing fuzzy systems from input-output data. Of these six methods, the LFE, MLFE, and CMs can be used to develop fuzzy systems from such data. The remaining three methods, RLS, BLS, and the GMs, can be used to take fuzzy systems that have been developed by the first group of methods and refine them with additional training data. Controlled Vocabulary Terms fuzzy systems; gradient methods; pattern clustering