Semi-Supervised Network Traffic Classification
Identifying and categorizing network traffic by application type is challenging because of the continued evolution of applications, especially of those with a desire to be undetectable. The diminished effectiveness of port-based identification and the overheads of deep packet inspection approaches motivate the people to classify traffic by exploiting distinctive flow characteristics of applications when they communicate on a network. This paper proposes a traffic classification methodology that relies on using only flow statistics to classify traffic. The authors introduce a flexible mathematical framework that leverages both labeled and unlabeled flows.