Information granulation-based fuzzy inference systems realized with the aid of space optimization and genetic algorithms

This paper is concerned with the design of fuzzy inference systems. A hybrid identification method of fuzzy inference systems based on a space optimization algorithm (SOA) and information granulation (IG) is developed. Information granulation realized with the aid of C-Means clustering is used to determine the initial values of the apex parameters of membership function of the fuzzy model. The SOA is exploited here to carry out the parametric optimization (which concerns the apexes-modal values of the membership functions) as well as structural optimization (involving the number of input variables to be used, a specific subset of input variables, the number of membership functions, and a type of the polynomial of the local model). Two well-known datasets are included to evaluate the performance of fuzzy models. A comparative analysis demonstrates that the SOA results in the improved performance both in terms of the quality of fuzzy model and the required computing time. Experimental results also highlight the superiority of the proposed model over the existing fuzzy and neural models.