The Application of Classification Methods in Intrusion Detection System

摘要：

Because of the high percent of misinformation in intrusion Detection System, we use classification methods in it in this paper. We mainly introduce the KNN and SVM. Using them we can get ideal result in processing non-structural data. If the dimension of data is not high, KNN can get more ideal result;and SVM can be applicable to the data which dimension is very high. So we create a new model for intrusion detection system based KNN and SVM; and we set a threshold for dimension to decide whether using KNN or SVM. In the end, we make some experiments to compare our system with traditional intrusion detection system.

关键词： Intrusion detection K_Nearest neighbor (KNN) Support vector machine (SVM) Character selection

作者: CHEN Fen WU Shunxiang

作者单位: Department of Automation Xiamen University Fujian, Xiamen 361005, China Department of Automation s Xiamen University Fujian, Xiamen 361005, China

会议类型: 国际会议

会议名称: 第一届国际计算机新科技与教育学术会议(Proceedings of the First International Conference on Computer Science & Education ICCSE2006)

会议地点: 厦门

会议语种:英文

页码: 586-588

在线出版日期: 2006-07-27（万方平台首次上网日期，不代表论文的发表时间）