【摘要】 研究网络安全问题,针对网络入侵数据是一种小样本、高维和冗余数据,传统检测方法无法进行很好降维且基于大样本数据,因此入侵检测率低。为了提高网络入侵检测率和网络安全,提出一种主成分分析(PCA)的支持向量机(SVM)网络入侵检测方法(PCA-SVM)。PCA-SVM首先通过PCA对网络入侵原始数据进行维数和消除冗余信息处理,减少了支持向量机的输入,采用粒子算法对支持向量机参数进行优化,获得最优网络入侵支持向量机检测模型,最后最优支持向量机模型对网络入侵数据进行测试。采用网络数据集在Matlab平台上对PCA-SVM算法进行仿真,结果表明,采用PCA-SVM加快了网络入侵检测速度,提高了检测率,降低了网络入侵漏报率,为网络入侵检测提供了一种实时检测工具。更多还原

【Abstract】 The problem of network security becomes more and more serious.Due to network intrusion data is a small sample,high peacekeeping redundant data;traditional detection methods intrusion detection rate is low.In order to improve the network intrusion detection rate and the network security,network intrusion detection method(PCA-SVM) is put forward.PCA-SVM uses the PCA to eliminate network intrusion original data dimension and redundant information,and then by using particle algorithm to optimize the support vector machines parameters,obtain the optimal network intrusion detection model,finally the optimal support vector machine(SVM) model is used to test the network intrusion data.The network intrusion data(KDD director 99) is used to test the PCA-SVM in Matlab,the results showed that,compared with other network intrusion detection methods,PCA-SVM accelerated network intrusion detection rate,increase the detection rate,reduce the network intrusion of fail.更多还原