A membership function selection method for fuzzy neural networks
Quingqing Zhou, Martin Purvis, Nikola Kirilov Kasabov · Otago University Research Archive (University of Otago) · 1997
Fuzzy neural networks provide for the extraction of fuzzy rules from artificial neural network architectures. In this paper we describe a general method, based on statistical analysis of the training data, for the selection of fuzzy membership functions to be used in connection with fuzzy neural networks. The technique is first described and then illustrated by means of two experimental examinations.