改进的WSVM入侵检测方法  

Network Intrusion Detection Based on Improved WSVM

作　　者：朱芳芳[1] 李志华[1] 王士同[1] 

机构地区：[1]江南大学信息工程学院,江苏无锡214122

出　　处：《计算机仿真》2008年第11期157-160,167,共5页Computer Simulation

摘　　要：在网络入侵检测中,数据类别不均衡训练集的使用将产生分类偏差。支持向量机(SVM)是一种新型的统计学习模型,在处理小样本和学习机的推广能力上有很大的优势。将加权支持向量机(WSVM)用于网络入侵检测中,使得入侵检测在小样本的条件下仍然具有较好的推广能力,并结合同时处理KDD Cup 1999 Data中的连续数据和离散字符,计算机仿真结果与其他方法比较结果表明,改进的WSVM入侵检测方法不但检测率提高,而且训练时间也缩短了,所以此改进方法是可行而高效的。In the network intrusion detection, the use of training sets with uneven class sizes will result in classification biases towards the class with larger training sizes. Support vector machine is a new statistical learning model, possesses great advantages in small sample and machine generalization ability. This paper introduces an improved weighted svm to network intrusion detection, and applies it to treat the continuous and discrete data together with KDD Cup 1999 Data,and comparison of detection ability between the above detection method and others is given. It is found that the the improved WSVM based network intrusion detection can shorten the training time, and is feasible and efficient.

关 键 词：支持向量机 入侵检测 加权系数 离散 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]