Identification the bast algorithm and features for skype traffic classification

Abstract

Skype uses strong encryption to secure communications inside the whole Skype network. Clients choose communication ports randomly. Therefore traditional port based or payload based identification of Skype traffic is not feasible. In this project we used a Machine Learning identification method to discover Skype host and voice calls as well. In this method, we test the whole algorithms in Weka application with five groups of features to show the most effective features and algorithm for Skype classification. Results indicate the Random forest and REPtree based approach perform much better than other algorithms on the identification of Skype traffic with accuracy 96.90% and 95.40% respectively.