Comparative study of type-2 fuzzy inference system optimization based on the uncertainty of membership functions

Abstract

A comparative study of type-2 fuzzy inference systems optimization as an integration method of Modular Neural Networks (MNNs) is presented. The optimization method for type-2 fuzzy systems is based on the footprint of uncertainty (FOU) of the membership functions. We use different benchmark problems to test the optimization method for the fuzzy systems. First, we tested the methodology by manually incrementing the percentage in the FOU, later we apply a Genetic Algorithm to find the optimal type-2 fuzzy system. We show the comparative results obtained for the benchmark problems. © 2010 Springer-Verlag Berlin Heidelberg.