Nonlinear Modeling Using Fuzzy Neural Networks Based on Scatter Space

In this paper, we introduce a design of fuzzy neural networks based on scatter space for nonlinear modeling. To design the networks, we partition the input space in the scatter form using fuzzy c-means (FCM) clustering algorithm which generates the fuzzy rules in the premise part of the proposed networks. The partitioned spaces express the fuzzy rules of the networks. Through this method, we are able to handle the high dimension problem. The consequence part of the rule is represented by polynomial functions whose coefficients are learned by standard back-propagation algorithm. The proposed networks are evaluated with the nonlinear process. Finally, this paper shows that the proposed networks can be utilized for high-dimension nonlinear process.