基于特征匹配和分箱技术的FCM算法研究

A FCM Algorithm Based on Character Matching and Binning

在基于聚类分析算法的入侵检测技术中,聚类的划分方法直接影响入侵检测的检测率。文章在基于分箱统计的HCM算法研究的基础上,针对模糊C-均值(FCM)算法的局限性,设计出一种改进的FCM算法。实验表明该算法比已有的FCM算法在对聚类的划分情况又有所改善,从而能提高检测率,且能较好地发现新的攻击类型。

In intrusion detection technology based on clustering analysis algorithm,the clustering＇s partition has a direct influence on the detection rate of the intrusion detection.In view of the limitations of the FCM clustering algorithm,the paper puts forward a modified FCM algorithm on the basis of the HCM algorithm based on binning.The experiment shows that such algorithm contributes to improve the clustering＇s partition and the detection rating,and have better access to the new attack.