在這項研究中，我們與台灣其中一個優秀的銀行Alpha合作，旨在通過因果提升建模技術提高數位行銷的效果。我們透過運用監督式機器學習技術和神經網路模型來解決優化客戶目標策略的挑戰。我們引入了兩階段收益提升模型的概念，並提出了使用神經網路模型進一步改進條件平均處理效果（CATE）估計的方法。我們展示了增益模型在預測客戶反應和優化行銷活動方面的有效性，通過分析客戶數據和進行實際實驗，從而為Alpha帶來了銷售和收益的增加。

In this study, we collaborate with Alpha, a leading bank in Taiwan, aiming to enhance the efficacy of digital marketing through causal uplift modeling techniques. We address the challenge of optimizing customer targeting strategies by employing supervised machine learning techniques and neural network models. We introduce the concept of the two-stage revenue uplift model and propose further advancements using neural network models to improve CATE estimation. By analyzing customer data and conducting field experiments, we demonstrate the effectiveness of uplift modeling in predicting customer responses and optimizing marketing campaigns, leading to increased sales and revenues for Alpha.

Ascarza, E. (2018). Retention futility: Targeting high-risk customers might be ineffective. Journal of marketing Research, 55(1), 80-98.
Baumann, A., Haupt, J., Gebert, F., & Lessmann, S. (2019). The price of privacy: An evaluation of the economic value of collecting clickstream data. Business & Information Systems Engineering, 61, 413-431.
De Caigny, A., Coussement, K., & De Bock, K. W. (2018). A new hybrid classification algorithm for customer churn prediction based on logistic regression and decision trees. European Journal of Operational Research, 269(2), 760-772.
Devriendt, F., Berrevoets, J., & Verbeke, W. (2021). Why you should stop predicting customer churn and start using uplift models. Information Sciences, 548, 497-515.
Gubela, R. M., & Lessmann, S. (2021). Uplift modeling with value-driven evaluation metrics. Decision Support Systems, 150, 113648.
Gubela, R. M., Lessmann, S., & Jaroszewicz, S. (2020). Response transformation and profit decomposition for revenue uplift modeling. European Journal of Operational Research, 283(2), 647-661.
Holland, P. W. (1986). Statistics and causal inference. Journal of the American statistical Association, 81(396), 945-960.
Künzel, S. R., Sekhon, J. S., Bickel, P. J., & Yu, B. (2019). Metalearners for estimating heterogeneous treatment effects using machine learning. Proceedings of the national academy of sciences, 116(10), 4156-4165.
Kane, K., Lo, V. S., & Zheng, J. (2014). Mining for the truly responsive customers and prospects using true-lift modeling: Comparison of new and existing methods. Journal of Marketing Analytics, 2, 218-238.
Olaya, D., Vásquez, J., Maldonado, S., Miranda, J., & Verbeke, W. (2020). Uplift Modeling for preventing student dropout in higher education. Decision Support Systems, 134, 113320.
Rubin, D. B. (1974). Estimating causal effects of treatments in randomized and nonrandomized studies. Journal of educational Psychology, 66(5), 688.
Rzepakowski, P., & Jaroszewicz, S. (2012). Decision trees for uplift modeling with single and multiple treatments. Knowledge and Information Systems, 32, 303-327.
Statista. (2023). Digital advertising spending worldwide from 2018 to 2028, by format (in billion U.S. dollars). In Statista; Statista Digital Market Insights. Statista.
Verbeke, W., Dejaeger, K., Martens, D., Hur, J., & Baesens, B. (2012). New insights into churn prediction in the telecommunication sector: A profit driven data mining approach. European Journal of Operational Research, 218(1), 211-229.
Zhan, B., Liu, C., Li, Y., & Wu, C. (2024). Weighted doubly robust learning: An uplift modeling technique for estimating mixed treatments' effect. Decision Support Systems, 176, 114060.