| 基于支持向量机和最小二乘支持向量机的入侵检测比较 |
| |
| 引用本文： | 任勋益,王汝传,谢永娟.基于支持向量机和最小二乘支持向量机的入侵检测比较[J].计算机科学,2008,35(10):83-85. |
| |
| 作者姓名： | 任勋益 王汝传 谢永娟 |
| |
| 作者单位： | 1. 南京邮电大学计算机学院,南京,210003
2. 南京邮电大学计算机学院,南京210003;南京大学计算机软件新技术国家重点实验室,南京210093 |
| |
| 基金项目： | 国家高技术研究发展计划(863计划)，江苏省高技术研究发展计划项目 |
| |
| 摘 要： | 将支持向量机和最小二乘支持向量机用于入侵检测之中,利用主元分析对数据进行约简,然后使用SVM和 LS-SVM对数据进行训练和测试.基于KDDCUP'99做了三组对比实验,对支持向量机和最小二乘支持向量机的性能做了统计.实验结果表明,SVM比LS-SVM分类能力强,但是LS-SVM耗时较少.
|
| 关 键 词： | 支持向量机 最小二乘支持向量机 入侵检测 主元分析 |
Comparisons of SVM and LS-SVM for Intrusion Detection |
| |
| REN Xun-yi,WANG Ru-chuan,XIE Yong-juan.Comparisons of SVM and LS-SVM for Intrusion Detection[J].Computer Science,2008,35(10):83-85. |
| |
| Authors: | REN Xun-yi WANG Ru-chuan XIE Yong-juan |
| |
| Affiliation: | REN Xun-yi1 WANG Ru-chuan1,2 XIE Yong-juan1(College of Computer,Nanjing University of Posts , Telecommunications,Nanjing 210003,China)1(State Key Laboratory for Novel Software Technology,Nanjing University,Nanjing 210093,China)2 |
| |
| Abstract: | This paper utilizes support vector machine and least square-support vector machine for intrusion detection.We normalizae data,reduce the data with principal component analysis,train and test reduced data with support vector and least square support vector machine.We do three experiments on KDDCUP'99 data set,and utilize Receiver Operating Characteristics curves to evaluate classifier's ability of SVM and LS-SVM,and statistic time cost.Experimental results show SVM has more classifying ability than LS-SVM,bu... |
| |
| Keywords: | Support vector machine Least square support vector machine Intrusion detection Principal component analysis |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《计算机科学》浏览原始摘要信息 |
|
点击此处可从《计算机科学》下载全文 |
|
