摘要
本文提出了基于CVM算法的入侵检测方法,这种方法先对样本集求其MEB问题,MEB问题的解就是决定分类超平面的支持矢量,然后再根据支持矢量的分布对网络的入侵行为进行分类。通过用KDD99数据的验证,证明了这种方法的有效性和可行性。
Intrusion Detection System based on core vector machine is presented in this paper. Firstly ,the minimum enclosing ball of a training data set is solved by MEB (Minimum Enclosing Ball) algorithm, then the optimal separating hyperplane is constructed by the solutions of the core sets. According to the distribution of the core sets, the different intrusion actions can be detected. The related experiment indicates that the algorithm is feasible and effective.
出处
《微计算机信息》
北大核心
2008年第18期45-46,24,共3页
Control & Automation
关键词
核心矢量机
核心集
入侵检测
Core Vector Machine
Core sets
Intrusion Detection System