Study on Process of Network Traffic Classification Using Machine Learning

Abstract

Classification of network traffic is the essential step for many network researches. Machine learning based approach is one of the most important approaches in the field of network traffic classification. Many related algorithms have been issued by researchers while the whole process contains a series of steps except building the algorithm and few researchers perform description. In this paper, a detailed workflow of machine learning based network traffic classification in campus network of SunYat-sen University is described, including steps such as data preparation and model construction. In the latter part of the paper, simple experiments are performed to prove the effectiveness of machine learning approach.