Self Learning Network Traffic Classification 英文文献资料.pdf

IEEE Sponsored 2nd International Conference on Innovations in Information Embedded and Communication Systems ICIIECS’15 Self Learning Network Traffic Classification Vandana M, Sruthy Manmadhan, Student, M.Tech Assistant Professor Computer Science & Engineering, Computer Science & Engineering NSS College of Engineering, Palakkad Kerala NSS College of Engineering, Palakkad, vandana19191@ Keralasruthym.88@ Abstract- Network management is part of traffic engineering the network managers must have a detailed knowledge about and security. The current solutions - Deep Packet Inspection applications and protocols. The user application may allow (DPI) and statistical classification, rely on the availability of a large delays or jitter, but the users might be very sensitive to training set. In case of these there is a cumbersome need to long wait times. Managing network traffic requires a regularly update the signatures. Further their visibility is judicious balance of all these priorities. limited to classes the classifier has been trained for. Unsupervised algorithms have been envisioned as a an Classification of traffic helps identify different alternative to automatically identify classes of traffic. applications and protocols that exist in a network. Different