Method and software for extracting fuzzy classification rules by subtractive clustering

Abstract

We present a fast and robust method for extracting fuzzy classification rules from data. The method uses subtractive clustering to obtain the initial rules; the rule parameters are then optimized by using an efficient gradient-descent algorithm. We also describe a user-friendly Apple Macintosh software package called RIFLEX (Rockwell International Fuzzy Logic EXplorer), in which this rule extraction method has been implemented.