A hybrid feature selection method based on fuzzy feature selection and consistency measures

In this paper, we present a new method for dealing with feature subset selection based on fuzzy methods and consistency measures for handling classification problems. In fuzzy classifier systems the classification is obtained by a number of fuzzy if-then rules including linguistic terms such as low and high that fuzzify each feature. First, we project the original data set into a fuzzy space, then we select the feature subset based on the consistency measures. The proposed method which is an integration of fuzzy feature subset selection and consistency measures can select relevant features to get higher average classification accuracy rates than each of the above mentioned methods. The applicability of the proposed method has been demonstrated by reducing the number of features used for the classification of nine real-world data sets.