一种基于特征敏感的信用卡欺诈检测模型
A Credit Card Fraud Detection Model Based on Feature Sensitivity
摘要: 金融欺诈交易检测一直是金融领域关注的重点问题。然而，现有的欺诈检测方法，在模型训练过程中忽略了对数据的特征敏感性做处理，导致对减少实际经济损失没有明显作用；因此提出了一种基于特征敏感的stacking集成学习方法FSBS (feature-sensitive based stacking)，首先选择几种不同质的基分类器当作第一层基准模型，然后通过交叉验证的方式得到概率形式的输出，最后通过一层特殊的stacking集成方法使模型对大金额交易样本有所偏置。最终实验证明，FSBS模型可以有效减少欺诈交易带来的经济损失。
Abstract: Financial fraud transaction detection has always been a key issue in the financial field. However, the existing fraud detection methods ignore the feature sensitivity of the data in the model training process, which has no obvious effect on reducing the actual economic loss; therefore, a feature-sensitive stacking integrated learning method FSBS ( feature-sensitive based stacking), firstly selects several different qualitative base classifiers as the first-level benchmark model, and then obtains the probabilistic output through cross-validation, and finally uses a special layer of stacking integration method to make the model for large amounts. The trading sample is biased. The final experiment proved that the FSBS model can effectively reduce the economic loss caused by fraudulent transactions.
文章引用：黄家元. 一种基于特征敏感的信用卡欺诈检测模型[J]. 计算机科学与应用, 2021, 11(12): 3019-3027. https://doi.org/10.12677/CSA.2021.1112305

参考文献

[1] 360企业安全. 2019网络诈骗趋势研究报告[R]. 2019.
[2] Ghosh, S. and Reilly, D.L. (1994) Credit Card Fraud Detection with a Neural-Network. Proceedings of the Twenty-Seventh Hawaii International Conference on System Sci-ences, Wailea, 4-7 January 1994, 621-630. [Google Scholar] [CrossRef] 
[3] Aleskerov, E., Freisleben, B. and Rao, B. (1997) CARDWATCH: A Neural Network Based Database Mining System for Credit Card Fraud Detection. Proceedings of the IEEE/IAFE 1997 Computational Intelligence for Financial Engineering (CIFEr), New York, 24-25 March 1997.
[4] Kolalikhormuji, M., Bazrafkan, M., Sharifian, M., et al. (2014) Credit Card Fraud Detection with a Cascade Artificial Neural Network and Imperialist Competitive Algorithm. International Journal of Computer Applications, 96, 1-9. [Google Scholar] [CrossRef] 
[5] 凌晨添. 进化神经网络在信用卡欺诈检测中的应用[J]. 微电子学与计算机, 2011, 28(10): 14-17.
[6] Maes, S., Tuyls, K., Vanschoenwinkel, B. and Manderick, B. (2002) Credit Card Fraud Detection Using Bayesian and Neural Networks. First International NAISO Congress on Neuro Fuzzy Technolo-gies, Havana, 16-19 January 2002, 1-8.
[7] Sahin, Y., Bulkan, S. and Duman, E. (2013) A Cost-Sensitive Decision Tree Approach for Fraud Detection. Expert Systems with Applications, 40, 5916-5923. [Google Scholar] [CrossRef] 
[8] Liu, C.W., et al. (2015) Financial Fraud Detection Model: Based on Random Forest. International Journal of Economics & Finance, 7, 178-188. [Google Scholar] [CrossRef] 
[9] Lu, H., Yang, L., Yan, K., et al. (2016) A Cost-Sensitive Rotation For-est Algorithm for Gene Expression Data Classification. Neurocomputing, 228, 270-276. [Google Scholar] [CrossRef] 
[10] Elkan, C. (2001) The Foundations of Cost-Sensitive Learning. In: Nebel, B., Ed., Proceedings of the 17th International Joint Conference on Artificial Intelligence, Morgan Kaufmann Publishers, Washington DC, 973-978.
[11] Zhang, Y. and Zhou, Z.H. (2010) Cost-Sensitive Face Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32, 1758-1769. [Google Scholar] [CrossRef] 
[12] Bnrescu, A. (2015) Detecting and Preventing Fraud with Data An-alytics. Procedia Economics and Finance, 32, 1827-1836. [Google Scholar] [CrossRef] 
[13] Alkhateeb, Z.K. and Maolood, A.T. (2019) Machine Learn-ing-Based Detection of Credit Card Fraud: A Comparative Study. American Journal of Engineering and Applied Sciences, 12, 535-542. [Google Scholar] [CrossRef] 
[14] Webga, K. and Lu, A. (2015) Discovery of Rating Fraud with Real-Time Streaming Visual Analytics. 2015 IEEE Symposium on Visualization for Cyber Security (VizSec), Chicago, 25 October 2015. [Google Scholar] [CrossRef] 
[15] Micenkova, B., McWilliams, B. and Assent, I. (2014) Learn-ing Representations for Outlier Detection on a Budget.