基于聚类融合的入侵检测  

An Intrusion Detection Based on Clustering Ensemble

作　　者：李建[1] 李杰[1] 孙燕花[1] 

机构地区：[1]中南大学信息科学与工程学院,湖南长沙410083

出　　处：《计算机技术与发展》2011年第10期250-252,F0003,共4页Computer Technology and Development

摘　　要：随着互联网的飞速发展,网络安全的问题日趋严重,传统的网络安全技术已难以应对日益繁多的网络攻击。因此入侵检测便应运而生了,而且其重要性日益提高。基于聚类分析的入侵检测已经成为其主要研究方向。聚类分析是一种有效的异常入侵检测方法,可用以在网络数据集中区分正常流量和异常流量。但单一的聚类算法很难达到预期的效果,为了提高入侵检测的效果,文中采用聚类融合技术,提出一种基于Co-assocition的模糊聚类融合算法,通过实验检测能显著提高检测率和降低误报率。With the rapid development of network, more and more network security problems are appearing, the traditional network security technology has been difficult to protect the network by growing range of network attacks. So the intrusion detection is turned out, and it gets more important in the network. Intrusion detection based on cluster analysis has become the main research directions. Cluster analysis is an effective method for anomaly intrusion detection, and it can distinguish the normal and abnormal data of the network data. But a single clustering algorithm is hard to achieve the desired effect. In order to improve the effectiveness of intrusion detection, proposes a new fuzzy clustering ensemble algorithm based On Co-assocition . Through experimental testing can significantly improve the detection rate and lower false alarm rate.

关 键 词：网络安全 入侵检测 聚类融合 

分 类 号：TP393.08[自动化与计算机技术—计算机应用技术]