Communication protocol identification based on data mining and automatic reasoning
Le Cai, Rong Shi, Xu Du · 2017
In recent years, big data has become an important resource of the information society, and the focus of research in all walks of life is to extract the effective information from big data with the greatest possibility. On the other hand, with the increase of the complexity of network data, the problem of cyber security becomes more and more serious. Protocol identification technology is an effective approach to solve this problem. Therefore in this paper we combine above two kinds of requirements together, and establish a system of protocol identification based on data mining and automatic reasoning. The inputs of this system are the massive bit streams in the network, and the data features are extracted by the data mining algorithm. Then it combines automatic classification with the automatic reasoning for protocol identification. The experiments show that the system not only can extract the feature information from the massive data, but also identify the network protocol effectively. It is an important basis for intrusion detection, traffic inspecting and other network security applications.