A Novel Feature Selection Method Based on Principal Component Analysis for Network Traffic Classification

Abstract:

In network traffic classification, by conventional PCA method, more features still exist due to uniform contribution rates for most of features. To overcome this problem, in this paper, a novel feature selection method is proposed to reduce data dimension of network traffic. A contribution rate of various features in each component is calculated by a new weight criterion. A maxima-order principle is proposed to determine feature selection. Based on three multi-class classification methods, performance comparison is conducted by actual traffic data with 10-fold cross-validation. Experiment shows that the proposed method has higher classification accuracy than conventional PCA method.

• Periodical: Advanced Materials Research (Volumes 989-994)
• Pages: 4510-4513
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.989-994.4510
• Online since: July 2014
• Authors: Hong Zhi Wang*, Jian Ping Zhang, Zun Yi Shang
• Keywords: Feature Selection, Maxima-Order, Network Traffic, PCA

References

[1] Feihu Yang. Research on Feature Selection Approach and its Application in Network Traffic Identification[D]. Nanjing: Nanjing Communication University, (2012).

[2] He Deng. Research on Network Traffic Classification Based on Machine Learning Method[D]. Zhuzhou: Hunan Industrial University, (2009).

[3] Moore A W, Zuev D. Internet traffic classification using bayesian analysis techniques[C]. International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS). Alberta (Canada), 2005: 50-60.

DOI: 10.1145/1064212.1064220

[4] Yang Aimin, Zhou Yongmei, Deng He, et al. Method of feature generation and selection for network traffic classification[J]. Journal of Shandong University, 2010, 40(5): 1-7.

[5] Huang Junyi, Wu Jing, Zhang Hui. Analysis of Feature Selection Effect on IP Traffic Classification Algorithms[J]. Computer Engineering, 2010, 36(16): 68-70.

[6] Huang Hongyan. Lightweight Intrusion Detection System Based on Feature Selection. [D]. Baoding: Hebei University, (2009).

[7] Haitham A. Jamil, Roozbeh Zarei, Nadir O. Fadlelssied, et al. Analysis of Features Selection for P2P Traffic Detection Using Support Vector Machine[C]. International Conference of Information and Communication Technology. Bandung, 2013: 116-121.

DOI: 10.1109/icoict.2013.6574558

[8] A. W. Moore, D. Zuev. Internet traffic classification using bayesian analysis techniques[C]. Proceedings of the 2005 ACM SIGMETRICS international conference on Measurement and Modeling of Computer Systems. USA(New York), 2005: 50-60.

DOI: 10.1145/1064212.1064220

[9] Li Jun, Zhang Shunyi, Liu Shidong, et al. P2P traffic identification technique[C]. Computational Intelligence and Security, 2007 International Conference on. China(Harbin), 2007: 37-41.

DOI: 10.1109/cis.2007.81

[10] Yuexiang Yang, Rui Wang, Yang Liu, et al. Solving P2P traffic identification problems Via optimized support vector machines[C]. Computer Systems and Applications, 2007. AICCSA'07. IEEE/ACS International Conference on. Amman, 2007: 165-171.

DOI: 10.1109/aiccsa.2007.370879

[11] Pascoal C,  Rosario de Oliveira M, Valadas R, et al. Robust Feature Selection and Robust PCA for Internet Traffic Anomaly Detection [C]. INFOCOM, 2012 Proceedings IEEE. Orlando, 2012: 1755-1763.

DOI: 10.1109/infcom.2012.6195548

[12] Naikal N, Yang A Y, Sastry S S. Informative feature selection for object recognition via sparse PCA[C]. Computer Vision (ICCV), 2011 IEEE International Conference on. Barcelona, 2011: 818-825.

DOI: 10.1109/iccv.2011.6126321

[13] J Friedman. Another approach to polychotomous classification, Technical report [R] Stanford University, Department of Statistics, (1996).

[14] V N Vapnik. Statistical Learning Theory[M]. Wiley Inter-Science, 1998: 493-520.

[15] Ran Wang, Sam Kwong, Degang Chen. A new method for multi-class support vector machines by training least number of classifiers[C]. Proceedings of the 2011 International Conference on Machine Learning and Cybernetics. China(Guilin), 2011: 648-653.

DOI: 10.1109/icmlc.2011.6016830