Comparison of Selection Method of a Membership Function for Fuzzy Neural Networks

Fuzzy neural networks are learning machine that realize the parameters of a fuzzy system (i.e., fuzzy sets, fuzzy rules) by exploiting approximation techniques from neural networks. In this paper, we tend to illustrate a general methodology, based on statistical analysis of the training data, for the choice of fuzzy membership functions to be utilized in reference to fuzzy neural networks. Fuzzy neural networks give for the extraction of fuzzy rules for from artificial neural network architectures. First, the technique is represented and so illustrated utilizing two experimental examinations for determining the alternate approach of the fuzzy neural network.