A NOVEL HYBRID ALGORITHM BASED ON CROW SEARCH ALGORITHM AND WHALE OPTIMIZATION ALGORITHM FOR HIGH-DIMENSIONAL OPTIMIZATION AND FEATURE SELECTION

Abstract

Crow search algorithm (CSA) is a meta-heuristic algorithm that mainly solves optimization problems.The weaknesses of the original CSA were its slow convergence speeds and inefficient exploitation capacity.Hence, this paper proposed a novel hybrid based on CSA and whale optimization algorithm (WOA), which is called HCSWOA for high-dimensional optimization problems.The main idea is to integrate two different algorithms' strengths into a proposed algorithm that utilizes the exploration ability of CSA with the exploitation and convergence abilities of WOA.To enhance the performance of the original WOA and CSA, this study employed an adaptive inertia weight strategy to improve exploitation and exploration capacities and convergence speed.The proposed algorithm has been compared against the original CSA, WOA, Grey Wolf Optimizer (GWO), Dragonfly Algorithm (DA), Particle Swarm Optimization (PSO), Sine Cosine Algorithm (SCA), Ant Lion Optimization algorithm (ALO), and Differential Evolution (DE) by using twenty-three standard benchmark functions and a real-world engineering problem as feature selection.The proposed algorithms have been examined on eighteen UCI standard and two DNA microarray datasets.The experimental results have revealed that HCSWOA has comprehensive superiority in solving global optimization and feature selection problems, which proves the capability of the proposed algorithm in solving real-world engineering problems.