Abstract

The Mayfly algorithm is an optimization method that offers a powerful hybrid algorithm structure, based on the behavior of mayflies. It combines major advantages of particle swarm optimization, genetic algorithm, and firefly algorithm. Simulation experiments proved that it is capable of optimizing both the benchmark functions but not without notable limitations. Slow convergent rate, premature convergent, and potential imbalance between exploration and exploitation were among notable shortcomings, due to the random selection procedure used which allows the existing algorithm to exploit specific areas in the search space. This has made it difficult for the Mayfly algorithm to be used to solve high-dimensional problem spaces such as feature selection. In this study, the Mayfly algorithm is enhanced with the roulette wheel selection method which will replace the random selection method used in the existing Mayfly algorithm. Both the existing Mayfly algorithm and formulated enhanced Mayfly algorithm were used as feature selection on the face, iris, and fused face – iris recognition system in other to determine the effects of the roulette wheel selection method used in enhancing the existing Mayfly algorithm. Simulation experiments were carried out and the result showed high positive effects of the roulette wheel selection method as a good replacement for random selection used in the conventional Mayfly algorithm.