[1] Zhao Wei, He Jianmin, Wang Chunlin, Chen Jinbo, et al. Application of a cost-sensitive methodfor churn prediction in telecommunication industry [J]. Journal of Southeast University (English Edition), 2007, 23 (1): 135-138. [doi:10.3969/j.issn.1003-7985.2007.01.027]
Copy
Application of a cost-sensitive methodfor churn prediction in telecommunication industry(
)
)Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]
- Volumn:
- 23
- Issue:
- 2007 1
- Page:
- 135-138
- Research Field:
- Economy and Management
- Publishing date:
- 2007-03-30
Info
- Title:
- Application of a cost-sensitive methodfor churn prediction in telecommunication industry
- Author(s):
- Zhao Wei; He Jianmin; Wang Chunlin; Chen Jinbo
- School of Economics and Management, Southeast University, Nanjing 210096, China
- PACS:
- F626;TP391
- DOI:
- 10.3969/j.issn.1003-7985.2007.01.027
- Abstract:
- To deal with the data mining problem of asymmetry misclassification cost, an innovative churn prediction method is proposed based on existing churn prediction research.This method adjusts the misclassification cost based on the C4.5 decision tree as a baseline classifier, which can obtain the prediction model with a minimum error rate based on the assumption that all misclassifications have the same cost, to realize cost-sensitive learning.Results from customer data of a certain Chinese telecommunication company and the fact that the churners and the non-churners have different misclassification costs demonstrate that by altering the sampling ratio of churners and non-churners, this cost-sensitive learning method can considerably reduce the total misclassification cost produced by traditional classification methods.This method can also play an important role in promoting core competence of Chinese telecommunication industry.
References:
[1] Hung S Y, Yen C D, Wang H Y.Applying data mining to telecom churn management[J].Expert Systems with Applications, 2006, 31(3):515-524.
[2] Ma H M, Yin H B, Guo X.The mining model of defecting customers[J]. Journal of Huazhong University of Science and Technology: Nature Science Edition, 2003, 31(9):28-30.(in Chinese)
[3] Mozer M C.Predicting subscriber dissatisfaction and improving retention in the wireless telecommunications industry [J].IEEE Trans on Neural Networks, 2000, 11(3):690-696.
[4] Yao M, Shen B, Li M F.A kind of unusual customers recognition system based on multi-criteria neural network and CART in telecom system[J].Systems Engineering — Theory & Practice, 2004, 24(5):78-83.(in Chinese)
[5] Chan P, Stolfo S.Towards scalable learning with non-uniform class and cost distributions:a case study in credit card fraud detection[C]//Proceedings of the 4th International Conference on Knowledge Discovery and Data Mining.New York, USA, 1998:164-168.
[6] Liu Y, Yang Y, Carbonell J.Boosting to correct inductive bias in text classification[C]//Proceedings of the Eleventh International Conference on Information and Knowledge Management.Pittsburgh, USA, 2002:348-355.
[7] Foster D P, Stine R A.Variable selection in data mining:building a predictive model for bankruptcy[J].Journal of the American Statistical Association, 2004, 99(466):303-313.
[8] Zheng E H, Li P, Song Z H.SVM-based cost sensitive mining[J].Information and Control, 2006, 35(3):294-298.(in Chinese)
[9] Quinlan J R.C4.5:Programs for machine learning[M].San Mateo, CA:Morgan Kaufmann Publishers, 1993.
[10] Duda R O, Hart P E, Stork D G.Pattern classification[M].2nd ed.John Wiley and Sons, Inc., 2000.
[11] Viaene S, Dedene G.Cost-sensitive learning and decision making revisited[J].European Journal of Operational Research, 2005, 166(11): 212-220.
[12] Ciraco M, Rogalewski M, Weiss G.Improving classifier utility by altering the misclassification cost ratio[C]//ACM SIGKDD Workshop on Utility-Based Data Mining.Chicago, USA, 2005:46-52.
Memo
- Memo:
- Biographies: Zhao Wei(1980—), male, graduate;He Jianmin(corresponding author), male, professor, nj.jian@public1.ptt.js.cn.