A Hybrid Grey Wolf-Whale Optimization Model for Feature Selection

Feature selection is crucial in reducing data dimensionality, improving classification performance, and minimizing computational costs in machine learning. This paper proposes a hybrid approach combining Grey Wolf Optimizer, Whale Optimization Algorithm, and Harmony Search Algorithm to select relevant features efficiently. The novelty of this approach lies in its ability to balance exploration and exploitation, outperforming standalone metaheuristics. Experimental results on benchmark datasets, including COVID-19 data, demonstrate superior performance over existing methods.