A binary whale optimization algorithm with hyperbolic tangent fitness function for feature selection

Abstract

To overcome the curse of dimensionality problem, a binary variant of the whale optimization algorithm (bWOA) with V-shaped is proposed. A hyperbolic tangent function is employed as a fitness function for mapping the continuous values to binary ones. Feature selection (FS) has attracted much attention in recent years and played a critical role in dealing with high-dimensional problems and can be modeled as an optimization problem. Eleven datasets from UCI repository from various applications are used. During the experiments, the effectiveness of feature selection is tested via a different type of data and size of features in the generic dataset. Furthermore, Wilcoxons rank-sum nonparametric statistical test was carried out at 5% significance level to judge whether the results of the proposed algorithms differ from those of the other compared algorithms in a statistically significant way. The quantitative and qualitative results revealed that the proposed binary algorithm in the FS domain is capable of minimizing the number of selected features as well as maximizing the classification accuracy within an appropriate time.