Importance of membership functions: a comparative study ondifferent learning methods for fuzzy inference systems

摘要

This paper investigates different adaptive structures for fuzzyinference systems. We examine the effect of membership functions onreasoning process when the number of rules is fixed. Three commonly usedmembership function shapes have been employed in this study. It has beenshown that membership functions have the dominant effect on reasoningprocess rather than number of rules or inference mechanism. We compareour adaptive membership function scheme with two already proposed byothers