支持向量机在入侵检测系统中的应用

Application of Support Vector Machine in Intrusion Detection System

SVM(支持向量机)方法被看作是对传统学习分类方法的一个好的替代,特别在小样本、非线性情况下,具有较好的泛化性能.本文简要分析了当前的几种入侵检测方法,重点介绍了SVM的学习算法,提出了将SVM用于入侵检测系统的方法.通过Matlab仿真实验,结果表明,运用SVM方法检测入侵,可以达到较高的准确检测率,是一种有效的入侵检测手段.

SVM(Support Vector Machine) approach is considered a good substitute for traditional learning classification, and has good generalization performance especially in small number and non-linear of training samples. This paper simply analyses several intrusion detecion methods, importantly introduces SVM’s learning algorithm, and suggests the approach using SVM in intrusion detection system. With Matlab simulation experiment, it can achieve higher accurate detection rate using SVM method, and is an effective i...