Abstract
This paper introduces new approaches to fuzzy inference system for system modeling based on input and output data using kernel machines. It is important issue how to select the best structure and parameters of fuzzy model from given input-output data. To solve this problem, this paper proposes the state-of-the-art kernel machine as the fuzzy inference engine. The kernel machine contains two modules such as the machine learning and the kernel function. The machine learning is a learning algorithm. The kernel function projects input data into high dimensional feature space. In this paper, the Support Vector Machine (SVM), Feature Vector Selection (FVS) and Relevance Vector Machine (RVM) as a kernel machine are presented. The proposed fuzzy system has the number of fuzzy rules and the parameter values of membership functions which are automatically generated. The structure of the proposed fuzzy inference system is equal to a Takagi-Sugeno fuzzy model. In addition, the number of fuzzy rules can be reduced under the condition of optimizing a given criterion by adjusting the linear transformation matrix or the parameter values of a kernel function using the Simulated results of the proposed technique are illustrated by examples involving benchmark nonlinear systems.
