基于大数据的K-means聚类算法的网络安全检测应用研究

Research on application of K-means clusteringalgorithm based on big data for network security detection

精准的入侵检测算法是确保网络安全的有效手段,为了解决传统检测算法准确性差、检测效率不高的问题,分析了网络入侵检测的现状,利用基于PReLU激活函数的ELM算法和改进的K-means算法设计了一种多级混合式入侵检测方法。利用NSL-KDD数据集进行算法检测效果的验证,结果表明:与传统的BP神经网络算法、SVM算法、ELM算法相比,多级混合式入侵检测方法检测率更高,精确度更高,并且大幅降低了误报率,在网络入侵类型判断方面具有更优异的检测效果。

Accurate intrusion detection algorithm is an effective means to ensure network security.In order to solve the problems of poor accuracy and low detection efficiency of traditional detection algorithms,this paper analyzes the current situation of network intrusion detection,and designs a multi-level hybrid intrusion detection method by using ELM algorithm based on PRELU activation function and improved K-means algorithm.The NSL-KDD data set is used to verify the detection effect of the algorithm.The results show that compared with the traditional BP neural network algorithm,SVM and ELM algorithm,the multi-level hybrid intrusion detection method has a higher detection rate,higher accuracy,and significantly reduces the false positive rate.It has a better detection effect in determining the network intrusion type.