基于客户细分和AdaBoost的电子商务客户流失预测研究

doi:10.3969/j.issn.1007-7375.e16-4206

摘要/Abstract

摘要： 为了准确识别高价值电子商务客户，提高对非流失客户的预测精度，本文首先对电子商务客户进行K-mediods聚类细分识别出高价值客户，再应用过采样和欠采样相结合的改进SMOTE处理不平衡的电子商务客户数据，最后用AdaBoost算法进行预测。实证研究表明，与成熟的客户流失预测算法BP神经网络、支持向量机（SVM）和改进支持向量机（CW-SVM）相比，该方法能更好地提高预测效果，与未细分前预测效果对比，客户细分后预测效果更好。

关键词: 客户细分, 不平衡数据, SMOTE算法, AdaBoost算法

Abstract: In order to identify the high value customers as well as improve the prediction accuracy of non-churn customers, the high value customer groups are identified by K-mediods clustering and the churn data processed with improved SMOTE (synthetic minority oversampling teachnique), which combines oversampling and undersampling methods to balance the datasets and generates the certain size of positive and negative samples by setting sampling ratio and controlling the model training time, then AdaBoost algorithm is employed to predict. At last, an empirical study on B2C E-commerce platform shows that the integrated model has better efficiency and higher prediction precision compared with the mature customer churn prediction algorithms, such as BP Neural, SVM (support vector machine) and CW-SVM (class weighted support vector machine). Meanwhile, the prediction model of e-commerce customer churning based on customer segmentation has been proved to have better prediction performance.

Key words: customer segmentation, imbalanced data, SMOTE, AdaBoost

中图分类号:

F270

武小军, 孟苏芳. 基于客户细分和AdaBoost的电子商务客户流失预测研究[J]. 工业工程, 2017, 20(2): 99-107.

WU Xiaojun, MENG Sufang. E-commerce Customer Churn Prediction based on Customer Segmentation and AdaBoost[J]. Industrial Engineering Journal , 2017, 20(2): 99-107.

参考文献

[1] RENJITH S. An integrated framework to recommend personalized retention actions to control B2C E-commerce customer churn [J]. International Journal of Engineering Trends and Technology, 2015, 27(3): 152-157.
[2] JU C H, LU Q B, GUO F P. E-commerce customer churn prediction model combined with individual activity [J]. Systems Engineering-Theory & Practice, 2013, 33(1): 141-150.
[3] 朱帮助. 基于SMC-RS-LSSVM的电子商务客户流失预测模型[J]. 系统工程理论与实践, 2010, 30(11): 1960-1967. ZHU Bangzhu. E-business customer churn prediction based on integration of SMC, rough sets and least square support vector machine [J]. Systems Engineering-Theory & Practice, 2010, 30(11): 1960-1967.
[4] YU X B, GUO S S, GUO J, et al. An extended support vector machine forecasting framework for customer churn in e-commerce [J]. Expert Systems with Applications, 2011, 38(3): 1425-1430.
[5] 朱帮助, 张秋菊, 邹昊飞, 等. 基于OSA算法和GMDH网络集成的电子商务客户流失预测[J]. 中国管理科学, 2011, 19(5): 64-70. ZHU Bangzhu, ZHANG Qiuju, ZOU Haofei, et al. E-business customer churn prediction based on integration of objective system analysis and group method of data handling network [J]. Chinese Journal of Management Science, 2011, 19(5): 64-70.
[6] SCHAPIRE R E. The strength of weak learnability [C]// Foundations of Computer Science, 1989. 30th Annual Symposium on. IEEE, 1989: 197-227.
[7] BREIMAN L. Bagging predictors [J]. Machine Learning, 1996, 24(2): 123-140.
[8] 应维云, 蔺楠, 谢雅雅, 等. 用LDA Boosting算法进行客户流失预测[J]. 数理统计与管理, 2010, 29(3): 400-408. YING Weiyun, LIN Nan, XIE Yiayia, et al. Research on the LDA boosting in customer churn prediction [J]. Journal of Applied Statistics & Management, 2010, 29(3): 400-408.
[9] LIU M, QIAO X Q, XU W L. Three categories customer churn prediction based on the adjusted real adaBoost [J]. Communication in Statistics-Simulation and Computation, 2011, 40(10): 1548-1562.
[10] 张玮, 杨善林, 刘婷婷. 基于CART和自适应Boosting算法的移动通信企业客户流失预测模型[J]. 中国管理科学, 2014, 22(10): 90-96. ZHANG Wei, YANG Shanlin, LIU Tingting. Customer churn prediction in mobile communication enterprises based on CART and Boosting algorithm [J]. Chinese Journal of Management Science, 2014, 22(10): 90-96.
[11] SUYKENS J A K, VANDEWALLE J. Least squares support vector machine classifiers [J]. Neural Processing Letters, 1999, 9(3): 293-300.
[12] 应维云, 覃正, 赵宇, 等. SVM方法及其在客户流失预测中的应用研究[J]. 系统工程理论与实践, 2007, 27(7): 105-110. YING Weiyun, QIN Zheng, ZHAO Yu, et al. Support vector machine and its application in customer churn prediction [J]. Systems Engineering--Theory & Practice, 2007, 27(7): 105-110.
[13] 赵铭, 李雪, 李秀婷, 等. 基于聚类分析的商业银行基金客户的分类研究[J]. 管理评论, 2013, 7: 38-44. ZHAO Ming, LI Xue, LI Xiuting, et al. Research on the classification of commercial banks' fund clients based on clustering analysis [J]. Management Review, 2013, 7: 38-44.
[14] 徐翔斌, 王佳强, 涂欢, 等. 基于改进RFM模型的电子商务客户细分[J]. 计算机应用, 2012, 32(5): 1439-1442. XU Xiangbin, WANG Jiaqiang, TU Huan, et al. Customer classification of E-commerce based on improved RFM model [J]. Journal of Computer Applications, 2012, 32(5): 1439-1442.
[15] BRITO P Q, SOARES C, ALMEIDA S, et al. Customer segmentation in a large database of an online customized fashion business [J]. Robotics and Computer-Integrated Manufacturing, 2015, 36: 93-100.
[16] VELMURUGAN T, SANTHANAM T. Computational complexity between K-means and K-medoids clustering algorithms for normal and uniform distributions of data points [J]. Journal of Computer Science, 2010, 6(3): 363-368.
[17] WANG D, THOMAS S G, WANG K L, et al. Finding groups in data, an itroduction to cluster analysis [J]. Applied Physics Letters, 1997, 70(12): 1593-1595.
[18] HADDEN J, TIWARI A, ROY R, et al. Computer assisted customer churn management: State-of-the-art and future trends [J]. Computers & Operations Research, 2007, 34(10): 2902-2917.
[19] CHAWLA N V, BOWYER K W, HALL L O, et al. SMOTE: synthetic minority over-sampling technique [J]. Journal of Artificial Intelligence Research, 2002, 16(1): 321-357.
[20] DRUMMOND C, HOLTE R C. C4. 5, class imbalance, and cost sensitivity: why under-sampling beats over-sampling [C]//Workshop on learning from imbalanced datasets II. 2003, 11.
[21] GANGANWAR V. An overview of classification algorithms for imbalanced datasets [J]. International Journal of Emerging Technology and Advanced Engineering, 2012, 2(4): 42-47.
[22] PROVOST F. Analysis and visualization of classifier performance: comparison under imprecise class and cost distributions[C]// International Conference on knowledge Discovery & Data Mining. 1999: 43-48.
[23] FAN X, TANG K. Enhanced maximum AUC linear classifier [C]// International Conference on Fuzzy Systems and Knowledge Discovery, Fskd 2010, 10-12 August 2010, Yantai, Shandong, China. 2010: 1540-1544.
[24] KUBAT M, MATWIN S. Addressing the curse of imbalanced training sets: one-sided selection [C]// In Proceedings of the Fourteenth International Conference on Machine Learning. 2000: 179--186.
[25] KAUFMAN L, ROUSSEEUW P J. Finding groups in data. an introduction to cluster analysis [J]. Wiley, 1990.