双支持向量机和聚类在入侵检测中的应用研究
pplying Twin Support Vector Machines and Clustering to Intrusin Detection
入侵检测系统正成为网络安全领域的一个关键部分。当前许多智能入侵检测模型的检测性能仍然不能满足真实网络环境的需要。本文提出了一个基于双支持向量机入侵检测模型，然后设计了一个基于聚类的样本降采样方法来从海量的审计数据中提取训练数据。使KDD'99 数据集来检验此入侵检测系统的检测性能，并与其他入侵检测系统作对比。实验结果表明，本文所提出的入侵检测系统在入侵检测率上取得了明显的进步。
Intrusion detection system (IDS) is becoming a critical component of network security. However, the detection rates of many proposed intelligent intrusion detection models are still not high enough to be applied to real network security. This paper presents a novel intrusion detection model with twin support vector machines. In addition, a k-means clustering sampler is designed to avoid the possible loss of significant data when sampled randomly. The performance of the proposed IDS is evaluated with KDD'99 dataset and compare to other intrusion detection models. The results demonstrate the superiority of the proposed IDS.
徐国胜、何俊
计算技术、计算机技术
网络安全双支持向量机K-means聚类入侵检测系统
Network securityTwin support vector machinesK-Means clusteringIntrusion detection
徐国胜,何俊.双支持向量机和聚类在入侵检测中的应用研究[EB/OL].(2014-05-15)[2025-08-05].http://www.paper.edu.cn/releasepaper/content/201405-236.点此复制
评论