• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2005, Vol. 27 ›› Issue (10): 5-7.

• 论文 • 上一篇    下一篇

基于数据挖掘的网络入侵检测中k-NN分类规则改进研究

李庆华 孟中楼 童健华   

  • 出版日期:2005-10-01 发布日期:2010-06-24

  • Online:2005-10-01 Published:2010-06-24

摘要:

采用数据挖掘技术来扩展入侵检测的功能以判别未知攻击是当前的一个研究热点。本文在分析了各种数据挖掘算法的基础上,提出将k-NN分类规则运用于入侵检测,给出了可 运用于入侵检测的k-NN分类规则改进算法k-NN for IDS。最后,我们在KDD99上对k-NN for IDS算法进行试验,验证了算法的有效性。

关键词: k-NN分类规则 入侵检测系统 规范化 信息增益 加权 k-NN for IDS

Abstract:

It is a hot research point to adopt data mining to expand intrusion detection systems' capability in order to detect new attacks. In this research,we apply the k_NN classifier to IDSs based on various data mining algorithms. We present k_NN for IDSs, an algorithm which is used for intrusion detection based on the k_NN classifier. Finally, we give a detailed description on using the k-NN classifier for IDSs, and prove the effectiveness of this mended  algorithm on KDD99.

Key words: (k-NN classifier, IDS, normalized, information gain, weighted, k-NN for IDS)