中文摘要：

为了提高网络入侵的检测率，以降低误检率，提出一种基于均值聚分析和多层核心集凝聚算法相融合的网络入侵检的网络入侵检测模型。利用K—means算法对多层核心集凝聚算法的核心集，用其替代原粗化过程得到的顶层核心集，实现了顶层核心集的快速准确定位，简化了算法的计算复杂性。然后，将KM．MulCA算法应用到入侵检测模型，最后采用KDDCup99数据集进行仿真实验。结果表明，本模型可以获得理想的网络入侵检测率和误检率。

英文摘要：

In order to improve the detection rate of intrusion detection model and reduce the false negative rate and error de- tection rate, this paper proposed a novel network intrusion detection model based on K-means and multilayer condensation algo- rithm. Firstly, it used K-means algorithm to obtain the core algorithm of MulCA set selection process, and set substitute for the top core raw coarsening process,realized the fast and accurate positioning of the core set, and might be appropriate to reduce the aggregation layer, simplified the computation complexity of the algorithm. And then, it applied the proposed algorithm to the intrusion detection model. The experimental results show that the proposed algorithm can obtain Rood intrusion results.