隨著顧客關係管理(Customer Relationship Management)的興起，越來越多 的業者希望透過資訊科技來瞭解並滿足客戶的需求，建立與現有顧客更緊密的關 係來提升顧客滿意度、降低顧客流失率，以提升企業的財務績效。因此，企業開 始嘗試以資料探勘來進行對顧客進行分析，譬如使用分群演算法將顧客分成不同 群體進行行銷，利用 BG/NBD 模型來預測顧客終生價值等。
現今，顧客流失預測模型為 CRM 系統中常見的模型之一，可告知管理者哪 些顧客將會離開企業，管理者即可針對此客群設計行銷活動等激勵措施，盡可能 地降低顧客實際的流失率。此模型已是企業管理顧客流失的基本方法之一，現在 常用於建立顧客模型的工具為羅吉斯迴歸(Logistic Regression)、類神經網路 (Neural Networks)、決策樹(Decision Tree)、支援向量機(Support Vector Machine, SVM)等。
本篇論文提供零售業的顧客流失預測模型，透過 DBSCAN 演算法將現有客群 的顧客交易紀錄依交易天數間隔進行分段，並以醫學存活分析研究中 Cox 比例風 險模型建立顧客流失預測模型，可依據顧客前六十天的資料，預測顧客未來是否 會流失。此模型可有效標記出可能流失的顧客，業者可針對這些顧客研究相關的 行銷策略。

With the growth of CRM, companies have started to meet customer needs through information technology, and establish a closer relationship with existing customers to reduce customer churn rate, in order to improve the company's financial performance. Therefore, companies began to use data mining to analyze customers, such as using clustering to divide customers into different cluster for marketing, and using BG/NBD models to predict customer lifetime value.
Customer churn problem was originally the main topic from telecom industry. The customer churn prediction model is the solution which can provide managers which customers will be lost, and design the campaign for these customers, and this model is already the basic solution for companies to manage customer churn. The tools commonly used to build customer models are Logistic Regression, Neural Networks, Decision Tree, and Support Vector Machines. (Support Vector Machine, SVM), etc.
In this paper, we provides a new method to segment customer transaction data of existing customer groups through the DBSCAN algorithm, and establish a churn prediction model based on the Cox proportional hazard model in medical survival analysis research, based on the previous 60 days of the customer Data to predict whether customers will be lost in the future.

[1] Falke Information, “Overview & Evolution of the Global Retail Industry”, 2016
[2] 行政院主計總處，國民所得統計及國內經濟情勢展望，民國109年
[3] Peppers D., and Rogers M., One to one B2B.,2001
[4] Berson et al., Building Data Mining Applications for CRM, McGraw-Hill, New York, NY,2000
[5] Strouse, Karen G, Marketing Telecommunications Services New Approaches for A Changing Environment, Artech House, Boston, 1999.
[6] Mou, Robb, and DeHoratius, “Retail store operations: Literature review and research directions”, European Journal of Operational Research, Vol 265, Issue 2, Pages 399-422, March 2018.
[7] Risselada, Hans, Peter C. Verhoef, and Tammo HA Bijmolt. "Staying power of churn prediction models." Journal of Interactive Marketing 24.3, Pages 198-208, 2010.
[8] Lu, J., Predicting customer churn in the telecommunications industry—an application of survival analysis modeling using SAS. ”SAS User Group International （SUGI27） Online Proceedings, pages. 114–127,2002
[9] Xia, Guo-en, and Wei-dong Jin. "Model of customer churn prediction on support vector machine." Systems Engineering-Theory & Practice 28.1, pages 71-77,2008.
[10] Sharma, Anuj & Panigrahi., “A Neural Network based Approach for Predicting Customer Churn in Cellular Network Services.” International Journal of Computer Applications. 27, 2013
[11] Kumar, Dudyala & Ravi, Vadlamani. “Predicting credit card customer churn in banks using data mining.” International Journal of Data Analysis Techniques and Strategies. 1. ,2008.
[12] Mattison, Rob, Telecom Churn Management-The Golden Opportunity, APDG Publishing, NC 27576, 2001.
[13] Zeynab Soltani and Nima Jafari Navimipour, “Customer relationship management mechanisms: A systematic review of the state of the art literature and reccomendations for future research”, Computers in Human Behavior, Vol 61, Pages 667-688, August 2016.
[14] V. Kumar, Denish Shah and Rajkumar Venkatesan, “Managing retailer profitability—one customer at a time.”, Journal of Retailing, Volume 82, Issue 4, Pages 277-294, 2006
[15] SAS Institute., “Best practice in churn prediction.”, A SAS Institute White Paper.,2000
[16] Wei, Chih-Ping, and I-Tang Chiu. "Turning telecommunications call details to churn prediction: a data mining approach.",Expert systems with applications Vol 23.2, Page 103-112 ,2002
[17] V. Nath, R.S. Behara, “Customer churn analysis in the wireless industry: a data mining approach”, Annual Meeting of the Decision Sciences Institute, Pages 505–510., 2003
[18] Tsai, Chih-Fong, and Yu-Hsin Lu. "Customer churn prediction by hybrid neural networks." Expert Systems with Applications, 36.10, page 12547-12553, 2009.
[19] S. Gürsoy, U. Tug ˘ba, “Customer churn analysis in telecommunication sector,” J. School Bus. Admin. Istanbul Univ, Vol.39, Pages 35–49.,2010
[20] SHAABAN, Essam, et al., “A proposed churn prediction model.”, International Journal of Engineering Research and Applications, 2.4: 693-697 2012.
[21] Naz et al., “a review on customer churn prediction data mining modeling techniques”. Indian Journal of Science and Technology., 2018
[22] Oyeniyi, A. O., Adeyemo, A. B., “Customer Churn Analysis In Banking Sector Using Data Mining Techniques.”, African Journal of Computing and ICT, 8（3）, Pages 165 - 174., 2015.
[23] GÜNTHER, Clara-Cecilie, et al., “Modelling and predicting customer churn from an insurance company.”, Scandinavian Actuarial Journal, Vol 2014.1, Pages 58-71.,2014
[24] Coussement, Kristof, and Koen W. De Bock. "Customer churn prediction in the online gambling industry: The beneficial effect of ensemble learning.", Journal of Business Research, 66.9, 1629-1636., 2013
[25] DATTA, Piew et al., “Automated cellular modeling and prediction on a large scale”. Artificial Intelligence Review, 2000
[26] C. Fraley and A. E. Raftery, “How Many Clusters? Which Clustering Method? Answers Via Model-Based Cluster Analysis”, Technical Report No. 329, Department of Statistics University of Washington, 1998.
[27] Patil, Annapurna P., et al. "Customer churn prediction for retail business." 2017 International Conference on Energy, Communication, Data Analytics and Soft Computing （ICECDS）. IEEE, 2017.
[28] Klein, John; Moeschberger, Melvin, Survival Analysis: Techniques for Censored and Truncated Data （Second ed.）, Springer, 2005
[29] Kalbfleisch JD, Prentice RL. The Statistical Analysis of Failure Time Data. 2nd ed., John Wiley and Sons, New York; 2011.
[30] Kaplan, Edward L., and Paul Meier. "Nonparametric estimation from incomplete observations." Journal of the American statistical association 53.282 （1958）: 457-481.
[31] Kleinbaum, David G. and Mitchel Klein. “Survival analysis”. Springer, 2012
[32] Hanley, James A. “The Breslow estimator of the nonparametric baseline survivor function in Cox’s regression model: some heuristics.”, 2008
[33] KIM, Julia. “Survival analysis”. Pediatrics in Review-Elk Grove, 2012
[34] Kapur K.C. and Lamberson L.R. “eliability in engineering design”, John Wiley & Sons, New York,1977
[35] Cao, Ricardo & Vilar, Juan & Devia Rivera, Andres E., “Modelling consumer credit risk via survival analysis.”, Statistics and Operations Research Transactions, Vol. 33, Nº. 1, 2009, pages. 3-30. 33. ,2009
[36] Lin, D. Y. "On the Breslow estimator." Lifetime data analysis 13.4, 471-480.,2007
[37] Breslow, Norman E. "Analysis of survival data under the proportional hazards model." International Statistical Review, 45-57, 1975
[38] Mandrekar, J.N. “Receiver Operating Characteristic Curve in Diagnostic Test Assessment”.,2010