SURVEY ON NETWORK TRAFFIC CLASSIFICATION TECHNIQUES WITH CORRELATION INFORMATION

Abstract

Network Traffic classification is the need of today’s emerging and rapidly growing computer network for
network traffic analysis, network management, security monitoring, flow detection, QoS and lawful interception. It is
possible to apply machine learning techniques to classify traffic based on flow statistical feature. Supervised and
unsupervised classification algorithms have been applied to classify traffic. Conventional methods for classification
include port based prediction and payload based deep inspection methods. In recent network environment the usual
methods undergoes from some troubles like dynamic ports and encrypted applications. The nearest neighbor (NN)
method having advantages, such as no need of training procedure, no risk of over fitting of parameters, and able to
handle a large number of classes. But, the performance of NN classifier method affected if the size of training data is
small. This paper conducts a survey on the various network traffic classification techniques, also focuses on a new non
parametric traffic classification method, which makes use of correlation information for the purpose of classification.