Classification Research on SSL Encrypted Application

Abstract

With the rapid development of computer and communication technology, people are increasingly dependent on computers, Internet and other infrastructures. At the same time, they also bear the risks and hazards of various kinds of security incidents. Secure Sockets Layer (SSL) is used to encrypt the information on the Internet so that data can be transferred safely. This paper focuses on the study of SSL encrypted traffic and Google is chosen as an example. First, some SSL encrypted applications are studied through SSL certificates, and then we apply the C4.5 machine learning algorithm to the classification of SSL encrypted applications, using packet length, packet inter-arrival time, and the direction of a flow as features. Our classification method yields a high precision and recall rate. © Springer-Verlag Berlin Heidelberg 2013.