北京邮电大学学报

  • EI核心期刊

北京邮电大学学报

• 论文 • 上一篇    下一篇

基于KFDA-SVM的入侵检测技术(增刊)

魏宇欣

  

  1. 北京邮电大学继续教育学院(即北京邮电大学通信网络综合技术研究所;北京邮电大学培训中心)
  • 收稿日期:2006-12-13 修回日期:2007-01-28 出版日期:2007-06-30 发布日期:2007-06-30
  • 通讯作者: 贾嘉

Intrusion Detection Technology Based on KFDA-SVM

  • Received:2006-12-13 Revised:2007-01-28 Online:2007-06-30 Published:2007-06-30

摘要: 为了提高分类正确率和减少训练时间,将特征抽取技术与分类算法结合,提出了一种基于KFDA-SVM的入侵检测技术。采用KFDA抽取最佳鉴别矢量,运用SVM对投影后的数据分类。同时根据入侵数据高维异构小样本的特性,提出一种基于HVDM的混和核函数。采用KDD 99数据集进行试验,验证了该算法的有效性。

关键词: 入侵检测, 核Fisher鉴别分析, 支持向量机, 混和核函数

Abstract: In order to improve the detection rate and reduce the training time, KFDA-SVM intrusion detection technology is proposed which combines the feature extraction technology and classification algorithm. In the proposed algorithm, the KFDA is used to extract the optimal discriminant vectors and then the SVM is adopted to classify the projected data. A mixture of kernels based HVDM is proposed according to the high dimensional and heterogeneous datasets acquired in the intrusion detection. Results of the experiment using KDD 99 indicate the effectiveness of the algorithm.

Key words: intrusion detection, kernel fisher discriminant analysis, support vector machine, mixture of kernels