摘要
提出了一种新型网络入侵检测分类模型,设计了一个基于支持向量机(SVM)的分类器。采用因子分析法(FA)将行为样本的众多相关网络特征融合成精简的综合特征,实现了对网络监测数据的降维。利用支持向量决策函数排序法(SVDFRM),通过支持决策向量函数得到网络行为的特征贡献率并提取网络行为的重要特征。KDD99数据集测试实验结果表明,提出的分类模型降维效果显著,具有较好的实时性和较高的检测率。
Presents a new network intrusion detection classification model and gives a support vector machine (SVM) based classifier is presented. A factor analysis (FA) algorithm is utilized to fuse numerous related network behaviors features into concise integrated features so as to reduce network data dimensions. A support vector decision function ranking method (SVDFRM) is used to calculate the contribution of network behaviors features, and then important network behaviors features are extracted. The experimental results demonstrate that this model has good dimension reduction performance and real time performance, and its detection rate is satisfying.
出处
《电子科技大学学报》
EI
CAS
CSCD
北大核心
2009年第2期240-244,共5页
Journal of University of Electronic Science and Technology of China
基金
国家自然科学基金(60776807)
国家863计划重点课题(2006AA12A106)
关键词
分类
因子分析法
网络入侵监测
支持向量决策函数排序法
支持向量机
classification
factor analysis
intrusion detection
support vector decision function ranking method
support vector machine