Abstract:
Feature selection (FS) is a pre-processing step that aims to eliminate the redundant and less-informative features to enhance the performance of data mining techniques. It is also considered as one of the key factors for improving classification problems in high-dimensional datasets. This paper proposes an efficient wrapper feature selection method based on Grey Wolf Optimizer(GWO). GWO is a recent metaheuristic algorithm that has been widely employed to solve diverse optimization problems. However, GWO mainly follows the search directions toward the leading wolves, making it prone to fall into local optima, especially when dealing with high-dimensional problems, which is the case when dealing with many biological datasets. An enhanced variation of GWO called EGWO, that adapts two enhancements, is introduced to overcome this specific shortcoming. In the first mode, a transition parameter is incorporated to move GWO from the exploration phase to exploitation phase. In addition, several adaptive non-linear decreasing formulas are introduced to control the transition parameters. In the second mode, a random-based search strategy is exploited to empower the diversity of the search process. Two binarization schemes using S-shaped and V-shaped transfer functions are incorporated to map the continuous search space into a binary one for dealing with FS problem. The efficiency of the proposed EGWO is validated on ten high-dimensional low-samples biological data. Our experimental results show promising performance of EGWO compared to the original GWO approach as well as other state-of-the-art techniques in terms of dimensionality reduction and the enhancement of classification performance.
