Compensatory Neurofuzzy Inference Systems for Pattern Classification

In this paper, a compensatory neurofuzzy inference system (CNIS) is proposed for classification applications. The compensatory-based fuzzy reasoning method using adaptive fuzzy operations of neurofuzzy inference systems makes fuzzy logic systems more adaptive and effective. Furthermore, an online learning algorithm is proposed to automatically construct the CNIS model. They are created and adapted as on-line learning proceeds via simultaneous structure and parameter learning. The structure learning is based on the fuzzy similarity measure and the parameter learning is based on back propagation algorithm. The simulation results have shown that 1) the CNIS model converges quickly, and 2) the CNIS model improves correct classification rates.