Methods and Applications of Abnormal Traffic Detection in Multi-cluster Networks
With the increase of network services and applications,the boundary of network traffic characteristics is blurred,and network malicious behaviors emerge in an endless stream,and the detection and analysis of network traffic anomalies has become the primary task to ensure network security.Based on the K-means algorithm,this paper explores and summarizes the methods and applications of network anomaly traffic detection,firstly,summarize the classification and common detection methods of network abnormal traffic.Secondly,explore the K-means method improved from three perspectives:selecting initial clustering centers,adding clustering evaluation functions,and other methods.From the innovation and limitations of the K-means algorithm based on Hadoop and Spark platforms,as well as the integration method of"K-means+others",summarize and propose solutions,Intended to contribute to the field of network abnormal traffic detection.
NETL
NSTL
万方数据
