In the generation of fuzzy systems

the primary work is to extract and modulate membership functions and fuzzy rules

but

using traditional methods

the amount of this work expands startlingly with the increasing of the number of variables.This paper presents a neural network based algorithm to automatically extract membership functions and rules of fuzzy systems.The complicated input-output relationship is firstly decomposed into the accumulation of simple input-output relationships.For each individual variable

the algorithm will generate a set of membership functions that are appropriate for all simple input-output relationships.The fuzzy rules of the whole system are then generated based on these membership functions.At last

an example is given to show the potential of the method.