Acomparative study on binary artificial bee colony optimization methods for feature selection

Feature selection is a major pre-processing technique which aims to pick distinctive features from a whole dataset. The major concern of the feature selection is to reduce the computational cost of the classification process; however it is also possible to improve the classification accuracy by eliminating irrelevant features. Artificial Bee Colony (ABC) algorithm is a bio-inspired swarm intelligence optimization method. It imitates food searching behaviors of honey bees. For continuous optimization problems, ABC produces fast and efficient solutions but it is not suitable for discrete problems such as feature selection. In this study, a comparative performance analysis of some binary variants of ABC is demonstrated on the feature selection problem to determine the best. Experimental results show that some versions of the binary ABC algorithm produce promising results.