Cultural algorithm for feature selection

In this paper a new cultural evolution based feature selection method is proposed. The present process contains a wrapper approach based on Cultural Genetic Algorithm (CGA) and naïve Bayes classifiers. Cultural evolutionary algorithms are used for searching the problem space to find all of the possible subsets of features and naïve Bayes classifier is employed to evaluate each subset of features. According to its fast convergence, CGA is expected to show higher performance compared with classical GA. The results show that the proposed approach outperforms GA on different datasets.