Abstract

Since the formulation of the Mayfly optimization algorithm in 2020, many researchers have proposed either improvement or used the conventional Mayfly optimization algorithm to proffer solutions to the optimization problems. In the Mayfly algorithm, the gravity coefficient which works similar to Particle Swarm Optimization’s inertia weight assists the achievement of a sufficient balance between exploration and exploitation. The gravity coefficient was fixed and gradually reduced over the iterations, allowing the existing Mayfly algorithm to exploit specific areas in the search space. This makes it difficult for the Mayfly algorithm to be used to solve high-dimensional problem spaces such as feature selection. In this paper, the new gravity coefficient widens the search space. The experimental result shows that the enhanced Mayfly algorithm (EMA) technique has ensured optimal computational efficiency in terms of its recognition accuracy, recognition time, false acceptance rate, and false rejection rate compared with the conventional Mayfly algorithm (MA). This result implies that the enhanced Mayfly algorithm would indeed increase the capability of the original Mayfly algorithm.