A SIMPLE BUT POWERFUL HEURISTIC METHOD FOR GENERATING FUZZY RULES FROM NUMERICAL DATA

In this paper, we propose a simple but powerful heuristic method for a utomatically generating fuzzy if-then rules from numerical data. Fuzzy if-then rules with nonfuzzy singletons (i.e., real numbers) in the co nsequent parts are generated by the proposed heuristic method. The mai n advantage of the proposed heuristic method is its simplicity, i.e., it involves neither time-consuming iterative learning procedures nor c omplicated rule generation mechanisms. We also suggest a linguistic re presentation method for deriving linguistic rules from fuzzy if-then r ules with consequent real numbers. The proposed linguistic approximati on method consists of two linguistic rule tables, which can realize ex actly the same nonlinear mapping as an original system based on fuzzy if-then rules with consequent real numbers. Using computer simulations on rice taste data, we demonstrate the high performance of the propos ed heuristic method and illustrate the proposed linguistic representat ion method. (C) 1997 Elsevier Science B.V.