Diagnosis and Diagnostic Rule Extraction Using Fuzzy Neural Network

Abstract

The goal of this study is to develop a reasoning device and a diagnostic rule extraction methodology based on fuzzy neural network. This paper describes a method to obtain a fuzzy neural network classifier from labeled training data sets and algorithms to extracted linguistic diagnostic rules from such a trained fuzzy neural network. Benchmark comparisons were performed using three data sets from three different fields of applications. The proposed methodology was shown to outperform all the existing methods that were compared.