BHHO-TVS: A Binary Harris Hawks Optimizer with Time-Varying Scheme for Solving Data Classification Problems

Data classification is a challenging problem. very sensitive to the noise and high dimensionality of data. Being able reduce model complexity can help improve accuracy performance. Therefore, in this research, we propose novel feature selection technique based on Binary Harris Hawks Optimizer with Time-Varying Scheme (BHHO-TVS). The proposed BHHO-TVS adopts time-varying transfer function that applied leverage influence location vector balance exploration exploitation power HHO. Eighteen well-known datasets provided by UCI repository were utilized show significance approach. reported results outperforms BHHO traditional binarization schemes as well other binary methods such gravitational search algorithm (BGSA), particle swarm optimization (BPSO), bat (BBA), whale (BWOA), salp (BSSA). Compared similar approaches introduced previous studies, method achieves best rates 67% datasets.