一种基于聚类算法的网络入侵检测应用

Network Intrusion Detection Based on Clustering Algorithm

针对网络入侵检测与聚类等问题,提出了一种综合模糊聚类与改进的SOM神经网络方法.通过对网络入侵数据提取、分析和处理,建立了网络入侵检测聚类模型,并对传统SOM网络层次进行改进,结合易发的网络入侵类型有针对性地对网络入侵数据进行聚类.网络入侵检测聚类与其他方法比较的结果表明,该模型在网络入侵检测聚类中具有更高的准确性和均衡性,该方法能有效提高网络入侵分类精度,减少聚类误差.

Aiming at easing the difficulty of intrusion detection and clustering for network, an integrated approach based on the fuzzy C-means （FCM） and self-organizing map （SOM） neural network was proposed in this paper, and the extraction, analysis and processing of network intrusion data done for accurately obtaining useful data. A network intrusion detection model was established, by which the traditional SOM network was updated to become a 3-layer network. According to network intrusion types, the network intrusion data was divided into five categories, and the network clustering result based on the integrated approach was made to compare with other methods. The results show that the model proposed in this paper has a higher accuracy and balance in the network intrusion detection clustering, and verified effectiveness and feasibility.