近年來隨著金融科技日新月異，支付行為也從傳統的實體貨幣，改為 線上支付，科技的發展提升消費者購物體驗以及企業競爭力，然而金融相關犯罪如駭客、個資剽竊也越來越常見，其中信用卡盜刷，是最常見、最麻煩的問題。
近期大數據分析、機器學習、演算法等領域的日漸成熟，使得銀行端與支付端對於盜刷行為有更好的預測能力。本文擬探討近年來信用卡防盜模式的建立過程及商業價值 過去在信用卡欺詐領域有很多研究 但是防盜專案 的建立非常複雜，每一個過程都有很多細節需要注意 最 困 難的部分是每天都有數以萬計的最新信用卡消費數據 當數據不是靜態分析而是動態分析時，會延伸出 許多問題降低模型的預測能力這便是資訊領域 中 的概念漂移。過去多數論文都 是以靜態資料為主，來研究不同模型和 演 算法之間的差異。本研究 以概念飄移現象為主軸，分析 專案流程中 各個階段常見的問題，並提出動態資料發生概念漂移的解決方法例如TWB移動窗格預測模型。透過流程觀的呈現與說明， 幫助非技術背景的管理者更理解信用卡防盜專案的內容，提升模型效果的同時，降低後續人力維護的成本，為企業帶來更多商業價值。

In recent years, with the rapid development of financial technology, payment behavior has changed from traditional physical currency to digital payment. However, technological development not only has a positive impact on society, financial-related crimes such as hacking and personal information plagiarism are becoming more and more common. Among them, credit card fraud is the most common and troublesome problem.
Recently, maturity of big data analysis, machine learning, algorithm and other fields has enabled banks to have a better ability to predict fraudulent behaviors. This paper intends to discuss the establishment process and commercial value of credit card fraud detection project in recent years. There have been many studies in the field of credit card fraud in the past, however, the establishment of the entire project is very complicated. There are many details to pay attention to in each process. The most difficult part is that there are tens of thousands of the latest credit card consumption data every day. When the data is not static analysis but dynamic, many problems will extend to reduce the predictive power of the model. This effect called concept drift in the information field. In the past, most of the papers focused on static data to study the differences between different models and algorithms. This research analyzes the common problems at each stage of the project process, and proposes a solution to the concept drift of dynamic data, helping managers with non-technical backgrounds better understand the content of the credit card fraud detection project.

卡優新聞網，2020。2019信用卡盜刷14億，超過9成來自網路交易。上網日期2021年5月20日。檢自:https://www.cardu.com.tw/news/detail.php?40676

詩伊，2018。行動支付？第三方支付？電子支付？別搞混了它們三個都不一樣! 上網日期2021年5月20日。檢自:https://agirls.aotter.net/post/54383

趨勢科技，2020。懷疑信用卡遭盜刷怎麼辦? 上網日期2021年5月22日。
檢自: https://blog.trendmicro.com.tw/?p=64761

CSDN博客，2016。壹讀: 增強學習、增量學習、遷移學習——概念性認知。上網日期2021年6月12日。檢自:
https://read01.com/zhtw/E7AeON.html#.YOgHZegzZPY

David Huang，2018。大鼻觀點: 不平衡資料的二元分類，選擇正確的衡量指標。上網日期2021年6月10日。檢自:
https://taweihuang.hpd.io/2018/12/28/imbalanced-data-performance-metrics/

Abdallah, A., M. A. Maarof and A. Zainal .2016. Fraud detection system: A survey.
Journal of Network and Computer Applications, 68: 90-113.

Douillard, A., Cord, M., Ollion, C., Robert, T., & Valle, E. 2020. Podnet: Pooled outputs distillation for small-tasks incremental learning. Computer vision-ECCV 2020-16th European conference, Glasgow, UK, August 23-28, 2020, Proceedings, Part XX.

Elkan, C. 2001. The foundations of cost-sensitive learning. Paper presented at the International joint conference on artificial intelligence.

Fu, K., D. Cheng, Y. Tu and L. Zhang. 2016. Credit card fraud detection using convolutional neural networks. International Conference on Neural Information Processing, Springer.

Fernández, A., S. García, M. Galar, R. C. Prati, B. Krawczyk and F. Herrera. 2018. Cost-sensitive learning. Learning from Imbalanced Data Sets, Springer: 63-78.

Hinton, G. E. and R. R. Salakhutdinov. 2006. Reducing the dimensionality of data with neural networks. science ,313(5786): 504-507.

Kaur, P., & Gosain, A. 2018. Comparing the behavior of oversampling and undersampling approach of class imbalance learning by combining class imbalance problem with noise. In ICT Based Innovations (pp. 23-30): Springer.

Mangala, D., & Kumari, P. 2015. Corporate fraud prevention and detection: Revisiting
the literature. Journal of Commerce & Accounting Research, 4(1), 35-45.

Makki, S., Assaghir, Z., Taher, Y., Haque, R., Hacid, M.-S., & Zeineddine, H. 2019. An experimental study with imbalanced classification approaches for credit card
fraud detection. IEEE Access, 7, 93010-93022.

Ma, T., S. Qian, J. Cao, G. Xue, J. Yu, Y. Zhu and M. Li. 2019. An Unsupervised Incremental Virtual Learning Method for Financial Fraud Detection. 2019 IEEE/ACS 16th International Conference on Computer Systems and Applications (AICCSA), IEEE.

Niu, X., L. Wang and X. Yang. 2019. A comparison study of credit card fraud detection: Supervised versus unsupervised. arXiv preprint arXiv: 1904.10604.

Şahin, Y. G. and E. Duman. 2011. Detecting credit card fraud by decision trees and support vector machines.

Shah, A. D., J. W. Bartlett, J. Carpenter, O. Nicholas and H. Hemingway. 2014.Comparison of random forest and parametric imputation models for imputing missing data using MICE: a CALIBER study. American journal of epidemiology, 179(6): 764-774.

Somasundaram, A. and S. Reddy. 2019. Parallel and incremental credit card fraud detection model to handle concept drift and data imbalance. Neural Computing and Applications, 31(1): 3-14.

Viegas, J. L., Cepeda, N. M., & Vieira, S. M. 2018. Electricity fraud detection using committee semi-supervised learning. International Joint Conference on Neural Networks (IJCNN).

West, J. and M. Bhattacharya. 2016. Intelligent financial fraud detection: a comprehensive review. Computers & security, 57: 47-66.