Whale Optimisation Algorithm for high-dimensional small-instance feature selection

Abstract

In this paper, eight variants of the Whale Optimisation Algorithm (WOA), that are based on eight different transfer functions, are introduced and used as search strategies in a wrapper feature selection model. Feature selection is a challenging task in machine learning process. It aims to minimise the size of a dataset by removing redundant and/or irrelevant features, with no information loss, to improve the efficiency of the learning algorithms. The used transfer functions belong to two different families; S-shaped and V-shaped. The proposed approaches have been tested on nine different high-dimensional medical datasets, with a low number of samples and multiple classes. The results revealed a superior performance for the V-shaped based approaches over the the S-shaped approaches. Moreover, the results of the V-shaped approach is compared with well-known feature selection approaches, and the superiority of the proposed approach is proven.

GRAPHICAL ABSTRACT

Keywords:

• Whale Optimisation Algorithm
• WOA
• transfer functions
• high-dimensional data
• classification
• feature selection

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work is supported by the research committee at Birzeit University [grant number 250177].