计算机与现代化

• 信息安全 • 上一篇    下一篇

基于组合算法选择特征的网络入侵检测模型

  

  1. (四川建筑职业技术学院网络管理中心,四川德阳618000)
  • 收稿日期:2014-03-31 出版日期:2014-08-15 发布日期:2014-08-19
  • 作者简介:刘春(1981-),男,四川内江人,四川建筑职业技术学院网络管理中心讲师,硕士,研究方向:计算机网络。

Network Intrusion Detection Model Based on Combination Algorithm Selecting Features

  1. (Network Management Center, Sichuan College of Architectural Technology, Deyang 618000, China)
  • Received:2014-03-31 Online:2014-08-15 Published:2014-08-19

摘要:

为了提高网络入侵检测的正确率,提出一种基于组合算法选择特征的网络入侵检测模型(GA-PSO)。首先建立网络入侵特征选择的
数学模型,采用遗传算法迅速找到网络入侵的特征子集,然后采用粒子群算法进一步选择,找到最优特征子集,最后采用极限学习机建
立网络入侵检测分类器,并采用KDD CUP 99数据集进行仿真测试。结果表明,GA-PSO不仅提高了入侵检测速度,而且可以提高网络入侵
检测的正确率。

关键词: 特征选择, 入侵检测, 遗传算法, 粒子群优化算法

Abstract:

In order to improve the detection accuracy of network intrusion, this paper proposed a network intrusion
detection model based on combination algorithm selecting features. Firstly, the mathematical model of network
intrusion detection features selecting problem is established, and then genetic algorithm is used to find the
feasible sub-features, and the optimal sub-features is obtained by particle swarm optimization algorithm, finally,
the network intrusion detection model is established by relevance vector machine, and the performance is test by
simulation experiments. The test results show that the proposed model can not only improved the detection speed,
but also can improve the network intrusion detection accuracy.

Key words:  , feature selection; intrusion detection; genetic algorithm; particle swarm optimization algorithm

中图分类号: