A Novel Hybrid WOA Algorithm enhanced with Evolutionary Strategy for High-Dimensional problems: An application on Microarray Cancer Data

Abstract

Abstract The stochastic optimization of continuous space for numerical problems has become a major challenge for scientists. The whale optimization algorithm (WOA) simulates the behavior of a humpback whale and is used to solve global optimization problems. Because of its excellent optimal performance and fewer required parameters, it has been widely accepted in a variety of engineering fields. Like other metaheuristics algorithms, WOA has the drawback of trapping in suboptimal regions and high dimensional ones. As a result, it is critical to examine the WOA components using powerful algorithms. A novel hybrid algorithm based on a recombinant evolutionary strategy is proposed to improve search capability. The developed method was analyzed using thirteen unconstrained benchmarked test functions. In addition, two data reduction techniques are used to overcome the dimensional curse. Meanwhile, the proposed algorithm was evaluated and contrasted on six microarray cancer datasets. The exhaustive examination and detailed results demonstrate that our new proposed structure has addressed main WOA’s shortcomings. Hence, a significant encouraging performance was observed of this newly developed RESHWOA algorithm.