Self-Learning Network Traffic Classification

Abstract

Network management is part of traffic engineering and security. The current solutions - Deep Packet Inspection (DPI) and statistical classification rely on the availability of a training set. In case of these there is a cumbersome need to regularly update the signatures. Further their visibility is limited to classes the classifier has been trained for. Unsupervised algorithms have been envisioned as an alternative to automatically identify classes of traffic. To address these issues Unsupervised Self Learning Network Traffic Classification is proposed. It uses unsupervised algorithms along with an adaptive seeding approach to automatically let classes of traffic to emerge, making them identified and labelled. Unlike traditional classifiers, there is no need of a-priori knowledge of neither signatures nor a training set to extract the signatures. Instead, Unsupervised Self Learning Network Traffic Classification automatically groups flows into pure (or homogeneous) clusters using simple statistical features. This label assignment (which is still based on some manual intervention) ensures that class labels can be easily discovered. Furthermore, Unsupervised Self Learning Network Traffic Classification uses an iterative seeding approach which will boost its ability to cope with new protocols and applications. Unlike state-of-art classifiers, the biggest advantage of Unsupervised Self Learning Network Traffic Classification is its ability to discover new protocols and applications in an almost automated fashion.

Authors and Affiliations

Vandana M, Sruthy Manmadhan

Keywords

Related Articles

Advanced Airport Baggage System

The main aim of the project is to design an “Advanced Airport Baggage System ” for aiding airports in managing the baggage handling system more efficiently. Current baggage transport is labor intensive and bears the ris...

A Highly Auto Configuration for A Node Using Addressing Protocol in Ad Hoc Networks

The addressing configuration may be a key challenge in networks owing to the shortage of infrastructure. Autonomous addressing protocols need a distributed and self-managed mechanism to avoid address collisions in an ex...

Use of Copper Tailings as the Partial Replacement of Sand in Concrete

In India, 4 million tons of copper tailings produce every year, out of which 25,000 tons produce in Khetri Copper Mines, Khetri, and Rajasthan. This research was undertaken to study the effects of copper mine tailings a...

 - Splitting Graphs

Let G(V,E) be a graph. A dominating set is a subset S of V such that every vertex not in S is adjacent to at least one vertex in S. The cardinality of a minimum dominating set is called the domination number, (G). A do...

A Review on Hardware Implementation for Reduction of Artifacts in Bio-Medical (ECG) Signal by Using Digital FIR-IIR Filter

ECG signal is of nature of time varying that is most common source used for the diagnosis purpose and observation and analysis of heart diseases present in the patient. ECG is recorded by placing electrodes at specified...

Download PDF file
  • EP ID EP21170
  • DOI -
  • Views 224
  • Downloads 5

How To Cite

Vandana M, Sruthy Manmadhan (2015). Self-Learning Network Traffic Classification. International Journal for Research in Applied Science and Engineering Technology (IJRASET), 3(8), -. https://europub.co.uk/articles/-A-21170