New fuzzy inference system using a support vector machine
Jong‐Cheol Kim, Sangchul Won · 2003
In this paper, we present a new support vector fuzzy inference system (SVFIS) for nonlinear system modeling. The proposed SVFIS is constructed using the support vector machine which does not have a bias term. The number of fuzzy rules is reduced by adjusting the parameter values of membership functions using the gradient descent method. Once a structure is selected, the parameter values in the consequent part of the Tagaki-Sugeno (TS) fuzzy model are determined by the least square method. The simulation result illustrates the effectiveness of the proposed SVFIS.