Optimizing the Performance of Machine Learning Based Traffic Classification

Abstract:

Traffic classification is a critical technology in the areas of network management and security monitoring. Traditional port-based and payload-based classification are no longer effective due to the fact that many applications utilize unpredictable port numbers and packet encryption. Researchers tend to apply machine learning (ML) techniques to identify the traffic flows by recognizing statistical features. Unfortunately, looking back upon the related work, most of the ML-based classification algorithms have similar performance, and what really matters now is how to optimize these techniques. In this paper, we analyzed two critical issues (Feature Selection, Configuration of Parameters) of ML classification, and presented the corresponding viable methods to optimize the classification model. This paper also reported the experimental evaluation to assess the performance improvements introduced by our optimized methods; experimental results on real-life datasets and network traffic show that the classification model successfully achieves significant accuracy improvement.

• Periodical: Advanced Materials Research (Volumes 756-759)
• Pages: 3506-3510
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.756-759.3506
• Online since: September 2013
• Authors: Qiu Chen Wang, Lei Wang, Ji Xiang
• Keywords: Machine Learning (ML), Optimize, Performance, Traffic Classification

References

[1] J. Frank: Proc. of the 17th National Computer Security Conference (1994), vol. 10.

[2] A. Moore and D. Zuev: ACM SIGMETRICS Performance Evaluation Review, Vol. 33 (2005) No. 1, p.50.

[3] T. Auld, A. Moore, and S. Gull: IEEE Transactions on Neural Networks, Vol. 18 (2007) No. 1, p.223.

[4] J. Park, H. Tyan, and C. Kuo: Proc. International Conference on Intelligent Information Hiding and Multimedia Signal Processing (2006), p.251.

[5] W. Lu, M. Tavallaee, and A. Ghorbani: Proc. of IEEE Global Telecommunications Conference (2009), p.1.

[6] N. Williams, S. Zander, and G. Armitage: ACM SIGCOMM Computer Communication Review, Vol. 36 (2006) No. 5, p.5.

[7] R. Yuan, Z. Li, X. Guan, and L. Xu: Information Systems Frontiers, Vol. 12 (2010) No. 2, p.149.

[8] Y. Lim, H. Kim, J. Jeong, C. Kim, T. Kwon, and Y. Choi: Proc. International Conference on Emerging Networking Experiments and Technologies (2010), p.9.

[9] Z. Liu and Q. Liu: Networks, IET, Vol. 1 (2012) No. 2, p.74.

[10] L. Bernaille, R. Teixeira, I. Akodkenou, A. Soule, and K. Salamatian: ACM SIGCOMM Computer Communication Review, Vol. 36 (2006) No. 2, p.23.

DOI: 10.1145/1129582.1129589

[11] M. Hall: Correlation-based Feature Selection for Machine Learning (Ph.D., The University of Waikato, New Zealand 1999).

[12] S. Russell, P. Norvig, J. Canny, J. Malik, and D. Edwards: Artificial Intelligence: A Modern Approach (Prentice hall Englewood Cliffs, 1995).

[13] N. Williams, S. Zander, and G. Armitage: Center for Advanced Internet Architectures, CA-IA, Technical Report B, Vol. 60410 (2006).

[14] Information on https: /www. cs. sfu. ca/people/Faculty/teaching/726/spring11/svmguide. pdf.