Improved intelligent dynamic swarm PSO algorithm and rough set for feature selection

Abstract

Feature Selection is one of the most important preprocessing steps in the field of Data Mining in handling dimensionality problems. It produces a smallest set of rules from the training data set with predetermined targets. Various techniques like Genetic Algorithm, Rough set, Swarm based approaches have been applied for Feature Selection (FS). Particle swarm Optimization was proved to be a competitive technique for FS. However it has certain limitations like premature convergence which is resolved by Intelligent Dynamic Swarm (IDS) algorithm. IDS could produce the reduct set in a smaller time complexity but lacks the accuracy. In this paper we propose an improvised algorithm of IDS for feature selection. © 2012 Springer-Verlag.