Fuzzy Classification Research Based On AFS Fuzzy Logic And Feature Selection

Since the theory of fuzzy sets and systems was proposed by an American cyberneticist L.A.Zadeh,in 1965,it has been studied widely and applied in many fields as a new mathematics method.AFS theory(Axiomatic Fuzzy Set) is a new method to study fuzzy set which was proposed by Professor Liu Xiaodong in 1995.It makes some of the mechanisms of decompose and composition of human conceptions to be understood with mathematical terms.It can be used to study the law of human thinking and be easily operated by computers.In essence,the AFS fuzzy system provides an effective tool to convert the information in the training examples into the membership functions and their fuzzy logic operations,and the membership functions and their logic operations are directly determined based on the distribution of original data,which are more rigorous and uniform.Now,AFS theory has been developed further and applied to fuzzy decision tree,credit rating analysis,pattern recognition and hitch diagnoses,etc.In many fuzzy systems,we often adopt the language labels(such as large,medium, small,etc.) to split the original feature into several fuzzy features.In order to reduce the computation complexity of the system after the fuzzification of features,the optimal fuzzy feature subset should be selected.This paper proposed a new heuristic algorithm Based on AFS Fuzzy Logic and fuzzy feature selection based on min-max learning rule and extension matrix to solve fuzzy Classification problem.First,split the original feature into several fuzzy features; second,employ a selection algorithm to select the fuzzy features,whose criterion and search strategy are min-max learning rule and fuzzy extension matrix;finally,gain the fuzzy classification association rules and obtain the fuzzy descriptions.Experimental results indicate it is a feasible algorithm which is easy and adaptable to the evaluation of the performance,and also have shown its good classification effect.