Semi-Supervised Machine Learning Algorithms for Classifying the Imbalanced Protocol Flows

Abstract:

The collision that between a car and a train is the main type of accidents in the case railway level-crossing, which is proved by the arrangement diagram analysing. The fault tree analysis and the event tree analysis are used to assess the level of the risk of the level-crossing quantificationally. Some conclusions can be drawn: the collision of the railway level-crossing that between a car and a train will happen 2.552 in a year, which can bring 0.061 equivalent fatalities. This paper puts forward some precautionary measures that based on the minimal cut set of the collision of the accident and the most probable or the highest risk event.

• Periodical: Applied Mechanics and Materials (Volumes 536-537)
• Pages: 809-814
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.536-537.809
• Online since: April 2014
• Authors: Lin Lin Li, Xiao Yi Zhang, Xia Zhang, Qing Li
• Keywords: Imbalance, Machine Learning, Semi-Supervised, Traffic Classification

References

[1] Wang Tao, Yu Shun-zheng. Advances in machine learing based network traffic calssification[J]. Journal of Chinese Computer Systems, 2012, 33(5): 1033-1040.

[2] Moore A, Zuev D. Internet traffic classification using Bayesian analysis techniques［C］. In ACM International Conference on Measurement and Modeling of Computer Systems ( SIGMETRICS) 2005，Banff, Alberta, Canada, June, 2005.

DOI: 10.1145/1064212.1064220

[3] Auld T，Moore A W，Gull S F. Bayesian neural networks for Internet traffic classification［J］. IEEE Trans. Neural Networks， January, 2007, 18(1): 223- 239.

DOI: 10.1109/tnn.2006.883010

[4] Xu Peng, Liu Qiong, Lin Sen. Internet traffic classification using support vector machine［J］. Journal of Computer Research and Development, 2009, 46(3): 407-414.

[5] Xu Peng, Lin Sen. Internet traffic classification using C4. 5 decision tree[J]. Journal of software, 2009, 20(10): 2692-2704.

DOI: 10.3724/sp.j.1001.2009.03444

[6] Zhou Jjianfeng, Yang Aimin, Liu Jicai. Traffic classification approach based on improved C4. 5 algorithm[J]. Computer Engineering and Application, 2012, 48(5): 71-74.

[7] Zhang Hong-li, Lu Gang. Machine learning algorithms for classifying the imbalanced protocol flows: evaluation and comparison[J]. Journal of software, 2012, 23(6): 1500 1516.

DOI: 10.3724/sp.j.1001.2012.04074

[8] Chen Yun-jing. P2P traffic identification technology research[D]. JiangSu: Yangzhou University，(2009).

[9] Erman J, Mahanti A, Arlit t M, et al. Offline/Online Traffic Classification Using Semi-supervised Learning[R]. Technical report. University of Calgary , (2007).

[10] Liu Bin, Li Zhi-tang, Tu Hao. Network application calssification method based on semi-supervised learning[J]. Microelectronics and Computer, 2008, 25(10): 113-116.

[11] Lin GZ, Xin Y, Niu XX, Jiang HB. Network traffic classification based on semi-supervised clustering[C]. The Journal of China Universities of Posts and Telecommunications, December 2010, 17(Suppl. 2): 84-88.

DOI: 10.1016/s1005-8885(09)60577-x

[12] Li X, Qi F, Xu D, et al. An internet traffic classification method based on semi-supervised support vector machine[C]/2011 IEEE International Conference on Communications(ICC).

DOI: 10.1109/icc.2011.5962736

[13] Zhang Zhen, Wang Bin-qiang, Li Xiang-tao, Huang Wan-wei. Semi-supervised traffic identification based on affinity propagation[J]. Acta automatica sinica, 2013, 39(7): 1100-1109.

DOI: 10.3724/sp.j.1004.2013.01100

[14] Zhang Yan, Lu Danju, Wu Baoguo. Research of Semi-supervised classification algorithm based on Tri-Training[J]. Computer technology and development, 2013, 23(7): 77-83.

[15] Moore A W, Zuev D. Discriminators for use in flow-based classification [R]. Cambridge: Intel Research, (2005).