《计算机应用研究》|Application Research of Computers

多粒度时序特征在离网预测中的应用

Application of multi-grain temporal features in churn prediction

免费全文下载 (已被下载 次)  
获取PDF全文
作者 石鸿斌,严建峰,白瑞瑞,徐彩旭,徐广根
机构 苏州大学 计算机科学与技术学院,江苏 苏州 215006
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)03-025-0767-05
DOI 10.19734/j.issn.1001-3695.2017.10.0987
摘要 电信运营商为了发现可能离网的客户,针对不同的场景研究开发了多种离网预测模型。目前的离网预测模型首先选择一种时间粒度抽取特征,之后使用机器学习算法对抽取的数据建模。这类方法只考虑了模型对分类性能的影响,没有充分考虑数据的作用。针对上述问题,提出一种使用多种时间粒度抽取特征的方法,并尝试在模型训练的不同阶段对不同粒度的特征进行融合。实验结果表明,使用多种粒度抽取特征训练出来的模型性能会明显优于使用单一粒度抽取特征的模型。
关键词 离网预测;时序数据;多粒度
基金项目 国家自然科学基金资助项目(61572339)
江苏省科技支撑计划重点项目(BE2014005)
本文URL http://www.arocmag.com/article/01-2019-03-025.html
英文标题 Application of multi-grain temporal features in churn prediction
作者英文名 Shi Hongbin, Yan Jianfeng, Bai Ruirui, Xu Caixu, Xu Guanggen
机构英文名 SchoolofComputer&Technology,SoochowUniversity,SuzhouJiangsu215006,China
英文摘要 Telecom operators have developed multiple churn prediction models to find potential users for different scenes.The present churn prediction models firstly select a kind of time granularity to extract features, then model the extracted data using machine learning algorithm.Such approaches only consider the influence of the model on classification performance, but the role of data is not fully considered.To solve this problem, this paper proposed a method which extracted multi-grain temporal features, and tried to integrate different granularity features at different training phases.Experimental results show that the performance of the model trained with multi-grain features is obviously superior than that trained with single granularity features.
英文关键词 churn prediction; time series data; multi-granularity
参考文献 查看稿件参考文献
  [1] Coussement K, Van den Poel D. Churn prediction in subscription services:an application of support vector machines while comparing two parameter-selection techniques[J] . Expert Systems with Applications, 2008, 34(1):313-327.
[2] Verbeke W, Dejaeger K, Martens D, et al. New insights into churn prediction in the telecommunication sector:a profit driven data mining approach[J] . European Journal of Operational Research, 2012, 218(1):211-229.
[3] Reinartz W J, Kumar V. The impact of customer relationship characteristics on profitable lifetime duration[J] . Journal of Marketing, 2003, 67(1):77-99.
[4] Guyon I, Lemaire V, Boullé M, et al. Design and analysis of the KDD cup 2009:fast scoring on a large orange customer database[J] . ACM SIGKDD Explorations Newsletter, 2010, 11(2):68-76.
[5] Yu H F, Lo H Y, Hsieh H P, et al. Feature engineering and classifier ensemble for KDD cup 2010[C] //Proc of JMLR:Workshop and Conference Proceedings. New York:ACM Press, 2010:1-16.
[6] Umayaparvathi V, Iyakutti K. Automated feature selection and churn prediction using deep learning models[J] . International Research Journal of Engineering and Technology, 2017, 4(3):1846-1854.
[7] Castanedo F, Valverde G, Zaratiegui J, et al. Using deep learning to predict customer churn in a mobile telecommunication network[EB/OL] . (2016-09-08). http://www. wiseathena. com/pdf/wa_dl. pdf.
[8] Wangperawong A, Brun C, Laudy O, et al. Churn analysis using deep convolutional neural networks and autoencoders[EB/OL] . (2016-04-20). https://arxiv. org/pdf/1604. 05377. pdf.
[9] Zaratiegui J, Montoro A, Castanedo F. Performing highly accurate predictions through convolutional networks for actual telecommunication challenges[EB/OL] . (2016-07-15). https://arxiv. org/abs/1511. 04906.
[10] Stripling E, Van den Broucke S, Antonio K, et al. Profit maximizing logistic regression modeling for customer churn prediction[C] //Proc of IEEE International Conference on Data Science and Advanced Analytics. Piscataway, NJ:IEEE Press, 2015:1-10.
[11] Lu Ning, Lin Hua, Lu Jie, et al. A customer churn prediction model in telecom industry using boosting[J] . IEEE Trans on Industrial Informatics, 2014, 10(2):1659-1665.
[12] Owczarczuk M. Churn models for prepaid customers in the cellular telecommunication industry using large data marts[J] . Expert Systems with Applications, 2010, 37(6):4710-4712.
[13] Zhao Xi, Shi Yong, Lee Jongwon, et al. Customer churn prediction based on feature clustering and nonparallel support vector machine[J] . International Journal of Information Technology & Decision Making, 2014, 13(5):1013-1027.
[14] Shaaban E, Helmy Y, Khedr A, et al. A proposed churn prediction model[J] . International Journal of Engineering Research and Applications, 2012, 2(4):693-697.
[15] Abbasimehr H, Setak M, Tarokh M J. A comparative assessment of the performance of ensemble learning in customer churn prediction[J] . Internal Arab Journal Information Technology, 2014, 11(6):599-606.
[16] Oseman K B, Haris N A, Bakar F A. Data mining in churn analysis model for telecommunication industry[J] . Journal of Statistical Modeling and Analytics, 2010, 2010(1):19-27.
[17] Kirui C, Hong Li, Cheruiyot W, et al. Predicting customer churn in mobile telephony industry using probabilistic classifiers in data mining[J] . International Journal of Computer Science Issues, 2013, 10(2):1694-0784.
[18] Idris A, Khan A. Ensemble based efficient churn prediction model for telecom[C] //Proc of the 12th International Conference on Frontiers of Information Technology. Piscataway, NJ:IEEE Press, 2014:238-244.
[19] Verbeke W, Martens D, Mues C, et al. Building comprehensible customer churn prediction models with advanced rule induction techniques[J] . Expert Systems with Applications, 2011, 38(3):2354-2364.
[20] Kim N, Jung K H, Kim Y S, et al. Uniformly subsampled ensemble(USE) for churn management:theory and implementation[J] . Expert Systems with Applications, 2012, 39(15):11839-11845.
[21] Wei C P, Chiu I T. Turning telecommunications call details to churn prediction:a data mining approach[J] . Expert Systems with Applications, 2002, 23(2):103-112.
[22] Zhang Junbo, Zheng Yu, Qi Dekang. Deep spatio-temporal residual networks for citywide crowd flows prediction[C] //Proc of the 31st AAAI Conference on Artificial Intelligence. San Francisco:AAAI Press, 2017.
[23] Breiman L. Random forests[J] . Machine Learning, 2001, 45(1):5-32.
[24] De Groot D S. Churn prediction in telecommunication[D] . Delft:Technische University Delft, 2017.
收稿日期 2017/10/29
修回日期 2017/12/15
页码 767-771
中图分类号 TP181
文献标志码 A