Machine Learned Real-Time Traffic Classifiers

Network traffic classification plays an important role in various network activities. Due to the ineffectiveness of traditional port-based and payload-based methods, recent works proposed using machine learning methods to classify flows based on statistical characteristics. In this study, we evaluate the effectiveness of machine learning techniques on the real-time traffic classification problem. We identify the most suitable ML classifier for network traffic classification by comparing various ML schemes,including both supervised and unsupervised methods. We also apply feature selection to identify significant features. Finally, we simulate real-time classification by using features derived from the first few packets of each flow.The results show that classifiers based on decision tree outperform others on both accuracy and performance; and that classifiers based on early flow properties can achieve high accuracy while reducing the computational complexity.