基于人工免疫网络和模糊C-均值聚类的入侵检测方法

摘要

The FCM clustering algorithm in intrusion detection method has two main shortcomings: sensitive to initial values and asking to input the number of clustering. In order to solve these two shortcomings, the intrusion detection method based on artificial immune network and fuzzy c-means algorithm is proposed by applying artificial immune network algorithm to FCM clustering algorithm. Through the simulation experiments on KDD_CUP99 data sets, the algorithm improves the detection rate and reduces the false alarm rate compared with the FCM algorithm. Experimental results show that this method can effectively detect intrusions in the networks.%针对入侵检测方法中模糊C-均值(FCM)聚类算法对初始值敏感和要求输入聚类数目的缺点,把人工免疫网络算法用于FCM聚类算法,提出了一种基于人工免疫网络和模糊C-均值的入侵检测方法.通过KDD_CUP1999数据集仿真试验,与FCM算法相比,该算法提高了检测率,降低了误警率.实验结果表明,该方法能够有效地检测网络中的入侵行为.