Using Machine Learning to Analyze Network Traffic Anomalies

In this paper, we study the application of machine learning methods, as well as spectral and statistical methods for real time network traffic anomaly detection. We determine the strengths and weaknesses of the existing methods and compare them in terms of efficiency. We then build a system of criteria to ensure the most efficient anomaly detection while meeting the specified system performance and resource consumption requirements. As a result, we suggest a set of the most effective anomaly detection methods as well as recommendations on the underlying system architecture.