| 基于混合比例方法抽样用于入侵检测 |
| |
| 引用本文： | 王庆荣,郑丽英. 基于混合比例方法抽样用于入侵检测[J]. 兰州交通大学学报, 2011, 30(1) |
| |
| 作者姓名： | 王庆荣 郑丽英 |
| |
| 作者单位： | 王庆荣,WANG Qing-rong(兰州交通大学,电子与信息工程学院,甘肃,兰州,730070;兰州理工大学,电气工程与信息工程学院,甘肃,兰州,730030);郑丽英,ZHENG Li-ying(兰州交通大学,电子与信息工程学院,甘肃,兰州,730070) |
| |
| 基金项目： | 教育部"春晖计划"项目，甘肃省自然科学基金 |
| |
| 摘 要： | 支持向量机对于样本量不足的情况下会使得入侵检测分类精度普遍下降.为了提高入侵检测的检测精度,提出了一种利用混合比例抽取样本研究不平衡数据集的方法.该方法首先将数据集按各种类别分开,然后不按比例抽取,使得每层样本中的数据量相对平衡.再用此样本集构造分类器,采用构造的分类器去构建入侵检测系统进行检测.实验结果表明,使用这种混合比例的不平衡数据集分类方法可以有效地提高检测精度,而且降低误报率.
|
| 关 键 词： | 不平衡数据集 支持向量机 入侵检测 |
Research on Intrusion Detection with the Method of Uixed Ration Sample |
| |
| WANG Qing-rong,ZHENG Li-ying. Research on Intrusion Detection with the Method of Uixed Ration Sample[J]. Journal of Lanzhou Jiaotong University, 2011, 30(1) |
| |
| Authors: | WANG Qing-rong ZHENG Li-ying |
| |
| Affiliation: | WANG Qing-rong1,2,ZHENG Li-ying1(1.School of Electronic and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China,2.School of Electronic and Information Enginneering,Lanzhou University of Science and Technology,Lanzhou 730030,China) |
| |
| Abstract: | Many studies show that support vector machines for insufficient sample can lead to the accuracy of intrusion detection decrease.In order to improve the accuracy of intrusion detection,a method is proposed which uses mixed ratio sample to extract sample and research unbalanced data set.This way firstly classifies the data set according to various categories,then extracts it out of proportion to keep relative balance to the date of each sample layer.At last it uses this sample to build classifier which is use... |
| |
| Keywords: | unbalanced data set support vector machine intrusion detection |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
