A METHOD FOR RULE REDUCTION IN A NEURO-FUZZY SYSTEM

摘要

This paper proposes a systematic procedure of rule reduction for fuzzy systems. The rule reduction problem in a neuro-fuzzy network is solved through an iterative algorithm aiming at selecting the minimal number of rules for the problem at hand. The reduction algorithm allows manipulation of a neuro-fuzzy system to minimize its complexity and to preserve its level of accuracy. Experimental results demonstrate the effectiveness of the algorithm in identifying reduced neuro-fuzzy systems with equivalent performance to the original ones.