Maned Wolf Optimizer: A Novel Nature-Inspired Metaheuristic Algorithm for Global Optimization and Feature Selection

Abstract

The increasing complexity of real-world optimization problems requires efficient and robust metaheuristic optimization algorithms. In this paper, a novel nature-inspired algorithm, called the Maned Wolf Optimizer (MWO), is proposed. The proposed algorithm is inspired by the hunting strategies and solitary behavior of maned wolves and is designed to achieve an effective balance between exploration and exploitation during the optimization process. The proposed algorithm was evaluated using 30 benchmark functions and applied to feature selection problems using eight public datasets. Experimental results demonstrate competitive performance against well-known metaheuristic algorithms, with MWO achieving an accuracy of 82.86% on the Sonar dataset, outperforming GWO, PSO, DE, and FEP.