Firefly optimization using artificial immune system for feature subset selection

Abstract

At first glance, the feature selection is a crucial step in a pattern recognition system. The main objective of this selection is to reduce the features number, by eliminating irrelevant and redundant attributes. In addition, we try to maintain or improve the classifier performance using neural network algorithm. Nevertheless, a new stochastic search strategy inspired by the clonal selection theory in an artificial immune system is proposed for feature subset selection. We have used the firefly and clonal selection algorithms to select the most relevant features in a dataset. In our proposed strategy, feature selection algorithm is formulated as an optimization problem that searches an optimum with less number of features in a feature space and a good accuracy. The goal of our study is to achieve a balance between the classification accuracy and the size of the feature subsets selected using two new hybrid algorithms based on Immune Firefly Algorithm (IFA). Our proposed approach has been evaluated on 10 standard datasets taken from UCI repository. The experimental outcomes have been compared to some popular feature selection methods. The comparison of results shows that our methods significantly outperform most of the used feature selection algorithms.