摘要
提出了一种基于SVM的网络异常流量检测新方法。分析了支持向量机的基本原理,结合网络流量异常检测的特点,讨论了异常检测的特征选择问题;提出了网络流量相关性、包长度统计变量以及异常报文统计等具有代表性的特征参数,描述了数据的预处理方法。试验结果表明,所选特征参数可有效地检测网络流量异常变化,说明基于支持向量机的在网络异常流量检测具有较好的可应用性。
An abnormal network traffic flow detection mechanism is presented based on support vector machine（SVM）. Theory of SVM is introduced first, and then the selection feature is discussed in depth Many features, including relativity of network traffic, length of packet and so on, are introduced in abnormal network traffic flow detection, and preprocessing of data is explained in detail. Experimental results show that the selected features can be used to detect the traffic anomaly incurred by network attacks and the detection mechanism based on SVM has the quality of generalization.
出处
《贵阳学院学报（自然科学版）》
2008年第1期23-26,共4页
Journal of Guiyang University：Natural Sciences
关键词
网络流量
支持向量机
网络安全
traffic flow
support vector machine
network security
