|
| 摘要: |
| 目前的入侵检测系统(IDS)存在着在先验知识较少的情况下推广能力差的问题。简述了IDS的基本原理，从本质上讲，入侵检测实际上是一个分类问题．就是通过检测把正常数据和异常数据分开。给出了入侵检测模型，论述了支持向量机(SVM)是在小样本学习的基础上发展起来的分类器设计方法．专门用于小样本数据．而且对数据维数不敏感．提出了基于SVM的通用入侵检测系统模型，它主要由审计数据预处理器、支持向量机分类器和决策系统3部分组成。说明了SVM系统模型的可行性、模型、工作过程、实现4方面的内容． |
| 关键词: 支持向量机 入侵检测系统 网络安全 统计学理论 |
| DOI： |
| 分类号:TP393.08 |
| 基金项目:国家自然科学基金;云南省重点实验室基金 |
|
| Application of support vector machine in intrusion detection system |
|
LING Yong-fa XIE Ji-ping
|
| Abstract: |
| The actual IDS(Intrusion Detection System) has poor expansion ability when there is less knowledge.The principle of IDS is introduced briefly.As an assortment in nature,IDS detaches the normal data from exceptional data by detection.The intrusion detection model is presented.The SVM(Support Vector Machine) is an assortment machine,which is specially design for small sample data and insensitive to data dimension.The general IDS model based on SVM is brought forward,which comprises three parts:audit data pretreatment processor,SVM assortment machine and decision-making system.Four aspects are focused on:feasibility,model structure,working process and imp-lementation. |
| Key words: support vector machine,intrusion detection system,networks security,statistical learning theory, |
