作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2013, Vol. 39 ›› Issue (5): 152-155. doi: 10.3969/j.issn.1000-3428.2013.05.033

• 安全技术 • 上一篇    下一篇

基于主成分分析的网络入侵检测算法

李占波1,白全海1,申义彩2   

  1. (1. 郑州大学信息工程学院,郑州 450000;2. 河南中医学院教务处,郑州 450000)
  • 收稿日期:2012-04-20 出版日期:2013-05-15 发布日期:2013-05-14
  • 作者简介:李占波(1965-),男,教授,主研方向:网络安全,电子政务,下一代互联网技术;白全海,硕士研究生;申义彩,讲师

Network Intrusion Detection Algorithm Based on Principal Component Analysis

LI Zhan-bo 1, BAI Quan-hai 1, SHEN Yi-cai 2   

  1. (1. College of Information Engineering, Zhengzhou University, Zhengzhou 450000, China; 2. Academic Affairs, Henan University of Traditional Chinese Medicine, Zhengzhou 450000, China)
  • Received:2012-04-20 Online:2013-05-15 Published:2013-05-14

摘要: 为提高入侵检测的效率和准确率,提出一种基于主成分分析法和K-最近邻算法的入侵检测算法。对原始攻击数据按其攻击类型的不同,分别利用主成分分析提取特征值,并通过K-最近邻算法对测试数据进行分类。Matlab仿真结果表明,将训练数据进行分类后再进行特征提取,能有效降低数据维数,提高分类算法的准确率。

关键词: 入侵检测算法, 主成分分析, K-最近邻算法, 特征值, 特征提取, 分类器

Abstract: To improve the efficiency and veracity of the intrusion detection, this paper presents an intrusion detection algorithm based on Principal Component Analysis(PCA) and K-nearest neighbor algorithm. This algorithm classifies the original attack data ordering by the class of attack, and extracts each class features based on the PCA. It uses the K-nearest neighbor algorithm to classify the observational data. Matlab simulations experiments result shows that this algorithm can effectively decrease the data dimension and enhance the veracity.

Key words: intrusion detection algorithm, Principal Component Analysis(PCA), K-nearest neighbor algorithm, feature value, feature extraction, classifier

中图分类号: