Fuzzy Modeling with Genetically Optimized Feature Space Reduction

This paper contributes to the ongoing studies on Genetic Algorithms applied to problems of feature selection in fuzzy modeling. The optimization scheme of Genetic Algorithms to reduce the dimensionality of input space is legitimate as the problem itself is of combinatorial nature. Fuzzy clustering realized through Fuzzy C-Means (FCM) is carried out in the reduced input space and the information granules obtained therein are used to form a series of local models of the rule-based fuzzy model. Our ultimate objective is to form a way of an efficient reduction of the input space leading to the enhanced interpretability of the fuzzy models and investigate possibilities of optimization of fuzzy models with respect. The experimental studies highlighting the numerical aspects of the design comprise synthetic data and data sets publicly available at several data sites.