DefinePK

Abstract

Network intrusion detection has been identified as one of the most challenging needs of the network security community in recent years. Intrusion detection systems (IDS) can analyze a large amount of data in a reasonable time to detect the attacks. Feature selection is necessary to reduce the time consumption and memory wastage. The dataset may be imprecise, incomplete or uncertain. Rough sets deals with vagueness and uncertainty. Rough set theory (RST) is used as a selection tool to find data dependencies and reduce the number of attributes which are redundant in a dataset. Particle swarm optimization (PSO) is known to effectively solve large-scale nonlinear optimization problems. An unsupervised hybrid feature selection based on PSO and RST for high dimensional network dataset is proposed. Feature selection algorithm namely PSO-quick reduct is applied for the different dimensions of network datasets. The simulation results for the unsupervised learning show that hybridization of PSO with rough set algorithm selects features more effectively than rough set algorithm without hybridization of PSO.