Performance evaluation of fuzzy classifier systems for multidimensional pattern classification problems

We examine the performance of a fuzzy genetics-based machine learning metho d for multidimensional pattern classification problems with continuous attr ibutes. In our method, each fuzzy if-then rule is handled as an individual, and a fitness value is assigned to each rule. Thus, our method can be view ed as a classifier system. In this paper, we first describe fuzzy if-then r ules and fuzzy reasoning for pattern classification problems. Then we expla in a genetics-based machine learning method that automatically generates fu zzy if-then rules for pattern classification problems from numerical data. Because our method uses linguistic values with fixed membership functions a s antecedent fuzzy sets, a linguistic interpretation of each fuzzy if-then rule is easily obtained. The fixed membership functions also lead to a simp le implementation of our method as a computer program. The simplicity of im plementation and the linguistic interpretation of the generated fuzzy if-th en rules are the main characteristic features of our method. The performanc e of our method is evaluated by computer simulations on some well-known tes t problems. While our method involves no tuning mechanism of membership fun ctions, it works very well in comparison with other classification methods such as nonfuzzy machine learning techniques and neural networks.