A Neuro-fuzzy Inference System WithImproved Performance
Spyros G. Tzafestaş, Giorgos B. Stamou · WIT transactions on information and communication technologies · 1970
Fuzzy systems and neural systems belong to the class of model free estimators and possess a high degree of parallelism. Due to these features several investigators have tried to produce several models of neuro-fuzzy inference systems with combined properties. The purpose of the present paper is to extend and improve one of these neuro-fuzzy structures such that to obtain better inferences. The system is based on the Keller-Yager-Tahani (K-Y-T) neuro-fuzzy model and uses Hamacher's intersection function /#(a,6) or Sugeno's complement function ĉ (a). The operation of the system is briefly described and it's features are established fn the form of four theorems. The capabilities of the system are shown by a set of simulation results derived for the case of trapezoidal fuzzy sets. These results are shown to be better than the ones obtained with the original neuro-fuzzy system of Keller, Yager and Tahant