摘要
在对现有的入侵检测技术研究的基础上,着重对数据挖掘技术中的聚类分析方法在入侵检测领域中的应用进行了研究。通过分析网络中数据的特点,提出了一种基于改进的k-means算法的无监督二次聚类算法,并用入侵检测权威数据集KDD Cup1999作为实验数据将其实现,实验表明,该算法具有较高的检测率和较低的误检率。
On the basis of the research on current intrusion detection technology, the application of clustering analysis in intrusion detection field is described. A twice clustering algorithm without supervising based on improved K-means algorithm is put forward by analyzing the data in the net. And the algorithm is actualized by using KDD Cup 1999 which is the authoritative data in intrusion detection. The experiment proves that the new algorithm has high right detecting rate and low error detecting rate.
出处
《河北省科学院学报》
CAS
2010年第3期31-34,共4页
Journal of The Hebei Academy of Sciences
关键词
入侵检测
数据挖掘
聚类分析
Intrusion detection
Data mining
Clustering analysis
