Network Traffic Analysis of Hierarchical Data Using Clustering

There is noteworthy enthusiasm for the information mining and network groups about the need to enhance existing methods for clustering multivariate network traffic stream records with the goal that we can rapidly derive basic traffic designs. In this venture, we explore the utilization of clustering procedures to distinguish intriguing traffic designs from network traffic information in a productive way. We build up a structure to manage blended sort qualities including numerical, categorical, and hierarchical attributes for a one-pass hierarchical clustering algorithm. In this network, we display a various leveled clustering plan for distinguishing noteworthy traffic stream designs. Specifically, we display a novel method for abusing the hierarchical structure of traffic attributes such as IP addresses, in combination with categorical and numerical attributes. This plan addresses the issues of network traffic investigation as it is a one-pass fixed memory clustering algorithm. We show the benefits of our clustering algorithm by producing a conservative report