A hybrid filter-based feature selection method via hesitant fuzzy and rough sets concept

High dimensional microarray datasets are difficult to classify since they have many features with small number of instances and imbalanced distribution of classes. This paper proposes a filter-based feature selection method to improve the classification performance of microarray datasets by selecting the significant features. Combining the concepts of rough sets, weighted rough set, fuzzy rough set and hesitant fuzzy sets for developing an effective algorithm is the main contribution of this paper. The mentioned method has two steps, in the first step, four discretization approaches are applied to discretize continuous datasets and selects a primary subset of features by combining of weighted rough set dependency degree and information gain via hesitant fuzzy aggregation approach. In the second step, a significance measure of features (defined by fuzzy rough concepts) is employed to remove redundant features from primary set. The Wilcoxon Signed Ranked tes (A Non-parametric statistical test) is conducted for comparing the presented method with ten feature selection methods across seven datasets. The results of experiments show that the proposed method is able to select a significant subset of features and it is an effective method in the literature in terms of classification performance and simplicity.