文献类型：期刊文献

英文题名：Network intrusion detection based on least squares support vector machine and chaos particle swarm optimization algorithm

作者：Zhang, Mohua[1]; Li, Ge[2]

第一作者：张墨华

通讯作者：Zhang, M.|[104844e872d56]张敏;[1048453a673f0]张曼;

机构：[1] School of Computer and Information Engineering, Henan University of Economics and Law, Zhengzhou 450002, China; [2] Department of Information Engineer, Conservancy Vocational Institute of North China Institute of Water Conservancy, Hydroelectric Power, Zhengzhou 450002, China

第一机构：河南财经政法大学计算机与信息工程学院

通讯机构：[1]School of Computer and Information Engineering, Henan University of Economics and Law, Zhengzhou 450002, China|[1048412]河南财经政法大学计算机与信息工程学院;[10484]河南财经政法大学;

年份：2012

卷号：7

期号：4

起止页码：169-174

外文期刊名：Journal of Convergence Information Technology

收录：EI(收录号：20121314903437);Scopus(收录号：2-s2.0-84863389244)

语种：英文

外文关键词：Chaotic systems - Least squares approximations - Network security - Particle swarm optimization (PSO) - Support vector machines - Vectors

摘要：Classification of intrusion attacks is a challenging problem in network security. In order to improve the classification accuracy of traditional intrusion detection methods, a novel intrusion detection technology by combining least squares support vector machine and chaos particle swarm optimization (CPSO-LSSVM) is proposed in the paper. Least squares support vector machine (LSSVM) is a popular pattern classification method with many diverse applications. However, the choice of the training parameters of least squares support vector machine has a heavy impact on its classification accuracy. Chaos particle swarm optimization is an evolutionary computation technique,which is better than particle swarm optimization algorithm.Thus, this study introduces CPSO as an optimization technique to simultaneously optimize the training parameter of LSSVM. The experimental results demonstrate that the proposed CPSO-LSSVM model has detection accuracy than other classifiers.