Classification Using an Efficient Neuro-Fuzzy Classifier Based on Adaptive Fuzzy Reasoning Method
Lin Cheng, Chun Cheng Peng · 2014
In this paper, a recurrent neuron-fuzzy classifier (RNFC) is proposed for use in classification applications. The compensatory fuzzy reasoning method uses adaptive fuzzy operations of neuro-fuzzy systems makes fuzzy logic systems more adaptive and effective. The recurrent network is embedded in the RNFC by adding feedback connections in the second layer, where the feedback units act as memory elements. Moreover, an online learning algorithm is proposed which can automatically construct the RNFC. There are no rules initially in the RNFC. They are created and adapted as online learning proceeds via simultaneous structure and parameter learning. Structure learning is based on the degree measure while parameter learning is based on the back propagation algorithm. The simulation results of the dynamic system modeling have shown that 1) the RNFC model converges quickly, and 2) the RNFC model improves correct classification rates.