计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (34): 113-115.DOI: 10.3778/j.issn.1002-8331.2008.34.035

• 网络、通信、安全 • 上一篇    下一篇

改进的SVM在入侵检测中的应用

童舜海   

  1. 丽水学院 计算机与信息工程学院,浙江 丽水 323000
  • 收稿日期:2007-12-19 修回日期:2008-03-07 出版日期:2008-12-01 发布日期:2008-12-01
  • 通讯作者: 童舜海

Application of improved support vector machines in intrusion detection

TONG Shun-hai   

  1. College of Computer and Information Engineering,Lishui University,Lishui,Zhejiang 323000,China
  • Received:2007-12-19 Revised:2008-03-07 Online:2008-12-01 Published:2008-12-01
  • Contact: TONG Shun-hai

摘要: 提出模糊支持向量机的入侵检测方法,根据输入样本对分类结果不同的影响程度,引入模糊隶属度,探讨了模糊支持向量(FSVM)原理。为进一步提高支持向量机的分类性能,提出Bagging算法对FSVM分类器进行集成,实验结果表明,提出的方法具有良好的检测性能。

关键词: 入侵检测, 支持向量机, 模糊隶属度, Bagging算法

Abstract: A intrusion detection based on fuzzy support vector machines is proposed.According to the different effects of input samples,the conception of the fuzzy membership to each input sample is considered and also the principle of Fuzzy Support Vector Machines(FSVM) is discussed.In order to improve the performance of FSVM classifier,Bagging algorithm is used to integrate FSVM.Then the evaluation method and results are given.The evaluation results show that the performance of the proposed algorithm is effective.

Key words: intrusion detection, support vector machines, fuzzy membership, Bagging algorithm