Gene subset selection using fuzzy statistical dependence technique and binary bat algorithm

The presence of big data may adversely affect obtaining classification accuracy in many life applications, such as genes dataset, which can contain many unnecessary data in the classification process. In this study, a two-stage mathematical model is proposed through which the features are selected. The first stage relies on the Fuzzy Statistical Dependence (FSD) technique, which is one of the filter techniques, and in the second stage, the Binary Bat Algorithm (BBA) is used, which depends on an appropriate fitness function to select important parameters. The experimental results proved that the proposed algorithm, which we refer to as FSD-BBA, excels over other methods in terms of classification accuracy and the number of influencing genes selected.