TIE is a project for application identification through network traffic analysis (aka Traffic Classification, Traffic Identification, etc.). We aim at building a common platform for the study and the development of traffic classification techniques by fostering collaboration among researchers and practitioners. TIE offers an open-source platform working as a multiple classifier system able to combine multiple classification techniques (implemented as separate plugins) and adopting different strategies of decision combination.
Scalable and Efficient Data Streaming Algorithms for Detecting Common Content in Internet Traffic. Minho Sung, Abhishek Kumar, Li Li, Jia Wang, Jun Xu. To appear in the Proc. of 2nd IEEE International Workshop on Networking Meets Databases (NetDB'06), April 2006. Sketch Guided Sampling -- Using On-Line Estimates of Flow Size for Adaptive Data Collection. Abhishek Kumar, Jun (Jim) Xu. To appear in the proceedings of IEEE Infocom'06, Barcelona, Spain, April 2006.