Non-Fuzzy Rule-Networks Based on Hard Clustering Algorithm

A new design of non-fuzzy-networks based on hard c-means (HCM) are introduced in this paper. To generate the ruels and design the networks, we use HCM clustering algorithm. The premise part of the rules of the proposed networks is expressed by the hard partition of input space generated by HCM clustering algorithm. The partitioned local spaces indicate the rules of the proposed networks. The consequence part of the rule is represented by polynomial functions. And back-propagation algorithm is used to learn the coefficients of the polynomial functions. The proposed networks are evaluated with the example for nonlinear process.