Development of an adaptive neuro-fuzzy classifier using linguistic hedges: Part 1

In this study, the development of an adaptive neuro-fuzzy classifier (ANFC) is proposed by using linguistic hedges (LHs). The LHs that are constituted by the power of fuzzy sets introduce the importance of the fuzzy sets for fuzzy rules. They can also change the primary meaning of fuzzy membership functions to secondary meaning. To improve the meaning of fuzzy rules and classification accuracy, a layer, which defines the adaptive linguistic hedges, is added into the proposed classifier network. The LHs are trained with other network parameters by scaled conjugate gradient (SCG) training algorithm. The tuned LH values of fuzzy sets improve the flexibility of fuzzy sets, this property of LH can improve the distinguishability rates of overlapped classes. The new classifier is compared with the other classifiers for different classification problems. The empirical results indicate that the recognition rates of the new classifier are better than the other fuzzy-based classification methods with less fuzzy rules.