Rough set based feature selection using swarm algorithms with improved initialization

Abstract

This paper is an attempt to improve the performance of swarm algorithms like Particle Swarm Optimization (PSO) and Intelligent Dynamic Swarm (IDS) through modifying the initialization. In this paper feature selection is done using Rough Set Theory (RST) and PSO or IDS. Improvement in the classification accuracy of reducts are observed when initialization is done using the proposed method. Statistical t-test are also performed for the validation of results. The fitness function used here is rough dependency measure. © 2018 American Scientific Publishers. All rights reserved.