A hybrid method for service identification of SSL/TLS encrypted traffic

It is significant and helpful to analyze encrypted traffic for solution of the potential network security problem. As one of the most popular encryption protocol, SSL/TLS has provided useful support for various network security while the network traffic has become complex and diverse. However, most of research on SSL/TLS traffic is about protocol level, identification and classification of services in real time is required for the effective network traffic management. In this paper, we propose a hybrid method to identify and classify the service in real time. We first filter the HTTPS traffic with a machine learning algorithm C4.5 decision tree, then classify the services with random forest, and process the traffic data what we get. Experimental results have shown that the accuracy reach more than 95%.