首页 >> 收录期刊 >> 软件学报 >> 正文
杂志中文名:软件学报
杂志英文名:Journal of Software
主管单位:中国科学院
主办单位:中国科学院软件研究所、中国计算机学会
地址:北京海淀区中关村南4街4号中科院软件所(8718信箱)
邮编:100080
电话:010-62562563 ;
Email:jos@iscas.ac.cn
ISSN:1000-9825
主编:李明树












基于支持向量机的入侵检测系统
引用本文:饶鲜,董春曦,杨绍全.基于支持向量机的入侵检测系统[J].软件学报,2003,14(4):798-803.
作者姓名:饶鲜  董春曦  杨绍全
作者单位:西安电子科技大学,电子工程系,电子对抗研究所,陕西,西安,710071
基金项目:Supported by the Military Communication Pre-Research Project of the 'Tenth Five-Year-Plan' of China under Grant No. 4100104030 ("十五"军事通讯预研)
摘    要:目前的入侵检测系统存在着在先验知识较少的情况下推广能力差的问题.在入侵检测系统中应用支持向量机算法,使得入侵检测系统在小样本(先验知识少)的条件下仍然具有良好的推广能力.首先介绍入侵检测研究的发展概况和支持向量机的分类算法,接着提出了基于支持向量机的入侵检测模型,然后以系统调用执行迹(system call trace)这类常用的入侵检测数据为例,详细讨论了该模型的工作过程,最后将计算机仿真结果与其他检测方法进行了比较.通过实验和比较发现,基于支持向量机的入侵检测系统不但所需要的先验知识远远小于其他方法,而且当检测性能相同时,该系统的训练时间将会缩短.

关 键 词:入侵检测  网络安全  支持向量机  统计学习  模式识别
修稿时间:8/2/2002 12:00:00 AM
作者简介:饶鲜(1976-),女,陕西城固人,博士生,讲师,主要研究领域为网络安全,信息对抗. Corresponding author: Phn: 86-29-8202274, E-mail: xianrao@yahoo.com.cn http://ecm.xidian.edu.cn

An Intrusion Detection System Based on Support Vector Machine
RAO Xian,DONG Chun-Xi and YANG Shao-Quan.An Intrusion Detection System Based on Support Vector Machine[J].Journal of Software,2003,14(4):798-803.
Authors:RAO Xian  DONG Chun-Xi  YANG Shao-Quan
Abstract:The generalizing ability of current IDS (intrusion detection system) is poor when given less priori knowledge. Utilizing SVM (support vector machines) in Intrusion Detection, the generalizing ability of IDS is still good when the sample size is small (less priori knowledge). First, the research progress of intrusion detection is recalled and algorithm of support vector machine taxonomy is introduced. Then the model of an Intrusion Detection System based on support vector machine is presented. An example using system call trace data, which is usually used in intrusion detection, is given to illustrate the performance of this model. Finally, comparison of detection ability between the above detection method and others is given. It is found that the IDS based on SVM needs less priori knowledge than other methods and can shorten the training time under the same detection performance condition.
Keywords:intrusion detection  network security  support vector machine  statistical learning  pattern recognition
本文献已被 CNKI 维普 万方数据 等数据库收录!
    浏览原始摘要     下载PDF全文