计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (30): 114-117.DOI: 10.3778/j.issn.1002-8331.2008.30.035

• 网络、通信、安全 • 上一篇    下一篇

基于BP神经网络和特征选择的入侵检测模型

吴 峻1,李 洋2   

  1. 1.湘潭大学,湖南 湘潭 411100
    2.中国科学院 计算技术研究所,北京 100080
  • 收稿日期:2007-11-28 修回日期:2008-03-18 出版日期:2008-10-21 发布日期:2008-10-21
  • 通讯作者: 吴 峻

Intrusion detection model based on BP neural network and feature selection

WU Jun1,LI Yang2   

  1. 1.Xiangtan University,Xiangtan,Hunan 411100,China
    2.Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100080,China
  • Received:2007-11-28 Revised:2008-03-18 Online:2008-10-21 Published:2008-10-21
  • Contact: WU Jun

摘要: 提出了一种基于后向传播神经网络和特征选择的入侵检测模型。通过使用该模型对经过特征提取后的攻击数据的训练学习,可以有效地识别各种入侵。在经典的KDD 1999数据集上的测试说明:该模型与传统的入侵检测模型相比,能够轻便、高效地对攻击模式进行训练学习,从而正确有效地检测网络攻击。

关键词: 网络安全, 入侵检测, 神经网络, 特征提取

Abstract: This paper proposes a kind of intrusion detection model based on Back Propagation(BP) neural network and feature selection mechanism.It can effectively detect several types of attacks after the process of feature selection and attack feature training.The experiments on classic KDD 1999 dataset demonstrate the model is accurate and effective.

Key words: network security, intrusion detection, neural network, feature selection