Survey of Traffic Classification using Machine Learning

Abstract

Traffic classification based on their generation applications has a very important role to play in network security and management.
Traditional methods include the port-based prediction methods and payload-based deep inspection methods. Within the current network environment,
the standard strategies suffer from variety of privacy issues, dynamic ports and encrypted applications. Recent research efforts are focused traffic
classification supported flow statistical options and Machine Learning Techniques. This paper conducts a survey on the various Machine Learning
(ML) techniques for IP traffic classification. Recent research tends to use machine learning techniques for classification.

Keywords: Traffic classification, Machine Learning (ML), Payload based-deep inspection methods.