Studying cost-sensitive learning for multi-class imbalance in Internet traffic classification

doi:10.1016/S1005-8885(11)60319-1

中国邮电高校学报(英文) ›› 2012, Vol. 19 ›› Issue (6): 63-72.doi: 10.1016/S1005-8885(11)60319-1

Studying cost-sensitive learning for multi-class imbalance in Internet traffic classification

刘珍¹,刘琼²

1. 华南理工大学
2. 华南理工大学软件学院

收稿日期:2012-05-03 修回日期:2012-09-23 出版日期:2012-12-31 发布日期:2012-12-14
通讯作者: 刘琼 E-mail:liuqiong@scut.edu.cn
基金资助:
This work was supported by the National Basic Research Program of China (2007CB07100, 2007CB07106).

Studying cost-sensitive learning for multi-class imbalance in Internet traffic classification

Zhen Liu¹,Qiong Liu

Received:2012-05-03 Revised:2012-09-23 Online:2012-12-31 Published:2012-12-14
Contact: Qiong Liu E-mail:liuqiong@scut.edu.cn
Supported by:
This work was supported by the National Basic Research Program of China (2007CB07100, 2007CB07106).

摘要/Abstract

摘要：

Cost-sensitive learning has been applied to resolve the multi-class imbalance problem in Internet traffic classification and it has achieved considerable results. But the classification performance on the minority classes with a few bytes is still unhopeful because the existing research only focuses on the classes with a large amount of bytes. Therefore, the class-dependent misclassification cost is studied. Firstly, the flow rate based cost matrix (FCM) is investigated. Secondly, a new cost matrix named weighted cost matrix (WCM) is proposed, which calculates a reasonable weight for each cost of FCM by regarding the data imbalance degree and classification accuracy of each class. It is able to further improve the classification performance on the difficult minority class (the class with more flows but worse classification accuracy). Experimental results on twelve real traffic datasets show that FCM and WCM obtain more than 92% flow g-mean and 80% byte g-mean on average; on the test set collected one year later, WCM outperforms FCM in terms of stability.

关键词:

Internet traffic classification, minority class, cost matrix, machine learning

Abstract:

Key words:

Internet traffic classification, minority class, cost matrix, machine learning

Zhen Liu Qiong Liu. Studying cost-sensitive learning for multi-class imbalance in Internet traffic classification[J]. Acta Metallurgica Sinica(English letters), 2012, 19(6): 63-72.

参考文献

1. Lin P, Yu X Y, Liu F, et al. A network traffic classification algorithm based on flow statistical characteristics. Journal of Beijing University of Posts and Telecommunications, 2008, 31(2): 15-19 (in Chinese)

2. Liu Q, Liu Z, Huang M. Study on Internet traffic classification using machine learning. Computer Science, 2010, 37(12): 35-39, 66 (in Chinese)

3. Moore A W, Zuev D. Internet traffic classification using Bayesian analysis techniques. Proceedings of the 2005 ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS’05), Jun 6-10, 2005, Banff, Canada. New York, NY, USA: ACM, 2005: 50-60

4. Jin Y, Duffield N, Erman J, et al. A modular machine learning system for flow-level traffic classification in large networks. ACM Transactions on Knowledge Discovery from Data, 2012 , 6(1): 1-34

5. Garcia V, Sanchez J S, Mollineda R A, et al. The class imbalance problem in pattern classification and learning. Proceedings of the 5th Spanish Workshop on Data Mining and Learning (TAMIDA’07), Sep, 2007, Zaragoza, Spain. 2007: 283-291

6. Domingos P. MetaCost: a general method for making classifiers cost-sensitive. Proceedings of the 5th ACM International Conference on Knowledge Discovery Data Mining (KDD’99), Aug 15-18, 1999. San Diego, CA, USA. New York, NY, USA: ACM, 1999: 155-164

7. Sun Y, Kamel M S, Wong A K C, et al. Cost-sensitive boosting for classification of imbalanced data. Pattern Recognition, 2007, 40(12): 3358-3378

8. He H B, Garcia E A. Learning from imbalanced data. IEEE Transactions on Knowledge and Data Engineering, 2009, 20(9): 1263-1284

9. Li Y F, Kwok J, Zhou Z H. Cost-sensitive semi-supervised support vector machine. Proceedings of the 24th AAAI Conference on Artificial Intelligence (AAAI’10),Jul 11-15, 2010,Atlanta, GA, USA. Menlo Park, CA, SA: AAAI, 2010: 500-505

10. Liu X Y, Zhou Z H. Towards cost-sensitive learning for real-world applications. Proceedings of the 15th International Conference on New Frontiers in Applied Data Mining (PAKDD’08), May 20-23, 2008, Osaka, Japan. New York, NY, USA: ACM, 2011: 494-505

11. Alejo R, Sotoca J M, Casan G A. An empirical study for the multi-class imbalance problem with neural networks. Proceedings of the 13th Iberoamerican Congress on Pattern Recognition (CIARP’08), Sep, 2008, La Havana, Cuba. LNCS 5197. Berlin, Germany: Springer-Verlag, 2008 : 479-486

12. Erman J, Mahanti A, Arlitt M. Byte me: a case for byte accuracy in traffic classification. Proceedings of the 3rd Annual ACM Workshop on Mining Network Data (MineNet’07), Jun 12-16, 2007,San Diego, CA, USA. New York, NY, USA: ACM, 2007: 35-38

13. He H T, Che C H, Ma F T, et al. Improve flow accuracy and byte accuracy in network traffic classification. Proceedings of the 4th International Conference on Intelligent Computing (ICIC’08), Sep 15-18, 2008, Shanghai, China.LNCS 5227. Berlin, Germany: Springer-Verlag, 2008: 449-458

14. Quinlan J R. C4.5: programs for machine learning. San Francisco, CA,USA: Morgan Kaufmann, 1993

15. Xu P, Lin S. Internet traffic classification using C4.5 decision tree. Journal of Software, 2009, 20(10): 2692-2704 (in Chinese)

16. Moore A W, Zuev D, Crogan M. Discriminators for use in flow-based classification.RR-05-13. Department of Computer Science, Queen Mary, University of London, 2005

17. Witten I H, Frank E. Data mining, practical machine learning tool and techniques. 2nd edition.San Francisco, CA, USA: Morgan Kaufmann, 2005, 403-418

Studying cost-sensitive learning for multi-class imbalance in Internet traffic classification

Studying cost-sensitive learning for multi-class imbalance in Internet traffic classification

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics

本文评价