基于数据增强与XGBoost 模型的信用卡欺诈检测算法

ASDS

Applied Statistics and Data Science

3066-84333066-8441

Art and Technology

10.61369/ASDS.2026040007

Article

基于数据增强与XGBoost 模型的信用卡欺诈检测算法https://artdesignp.com/journal/ASDS/2/4/10.61369/ASDS.2026040007邓磊

2026

2026-04-20

信用卡欺诈行为对经济、金融等诸多领域产生严重不良影响，如何有效预防、识别信用卡欺诈行为具有重要理论和实际价值。另一方面，来自银行匿名的真实数据集，高度不平衡，正常数据远远大于欺诈数据。本文首先基于CGAN 和Isolation Forest 对数据加强，构建贝叶斯优化器的XGBoost 模型，用于检测信用卡欺诈，称为CGAN-IF-Bayes-XGB 算法。具体来说使用CGAN 生成新的、真实的数据来补充原始数据集，Isolation Forest 则用于计算每条样本的“异常性”评分，作为新的特征注入，以此来增加样本中的有效信息量，为XGBoost 模型更好地识别诈骗交易做好铺垫。在CGAN-IF-Bayes-XGB 算法中使用了贝叶斯优化器来对XGBoost 中的超参数进行优化，确保XGBoost识别效果达到最佳。其次，本文把CGAN-IF-Bayes-XGB 算法与其他常用的机器学习算法（例如KNN 和Light GBM）进行了比较，比较结果表明：CGAN-IF-Bayes-XGB 算法具有更高的准确率、真阳性率、真阴性率和马太相关系数，这使其成为检测信用卡欺诈的有效方法。CGAN 模型,XGBoost 模型,Isolation Forest 模型,贝叶斯优化器

[1] Nassif., A.B.Talib., M.A.Nasir., Q.Dakalbab., F.M. Machine Learning for Anomaly Detection: A Systematic Review[J]. IEEE Access, 2021, 9,:pp.78658–78700.[2] THE NILSON REPORT. Global Card Fraud Losses at $33 Billion[R]. GlobeNewswire, 2026-01.[3] 人民网. 信用卡诈骗案一年5 万起银行损失数百亿仅追回16.5 亿元[N]. 2016-08.[4] SUBUDHI S., PANIGRAHI S., DASH S.R. ARMS: Automated rules management system for fraud detection[EB/OL]. (2020-02)[2026-01-31]. http://arxiv.org/abs/2002.06075.[5]LEE J., PARK K. GAN-based imbalanced data intrusion detection system[J]. Personal and Ubiquitous Computing, 2021, 25: 121-128.[6]GOODFELLOW I.J., POUGET-ABADIE J., MIRZA M., XU B., WARDE-FARLEY D., OZAIR S., COURVILLE A., BENGIO Y. Generative Adversarial Nets[C]//GHAHRAMANI Z., WELLING M., CORTES C., LAWRENCE N.D., WEINBERGER K.Q., eds. Advances in Neural Information Processing Systems 27. Montreal, Canada:Neural Information Processing Systems Foundation; Curran Associates, Inc., 2014: 2672-2680.[7] Liu F.T., Ting K.M., Zhou Z. Isolation-based anomaly detection ACM Trans[J]. Knowl. Discov. Data (TKDD), 2012, 6 (1): 1-39.[8]Li, X., Liang, D., Li, X. et al.. Quality monitoring of real-time PPP service using isolation forest-based residual anomaly detection[J]. GPS Solutions, 2024, 28(3): 118.[9]LIU F.T., TING K.M., ZHOU Z. Isolation Forest[J]. IEEE Transactions on Knowledge and Data Engineering, 2008, 20(8): 413-422.[10]SNOEK J., LAROCHELLE H., ADAMS R.P. Practical Bayesian Optimization of Machine Learning Algorithms[C]// LECUN Y., BENJIOU Y., HINTON G., eds. Advances in Neural Information Processing Systems 25. Lake Tahoe, NV, USA: Neural Information Processing Systems Foundation; Curran Associates, Inc., 2012: 2951-2959.[11]BERGSTRA J., YAMINS D., COX D.D. Making a Science of Model Search: Hyperparameter Optimization in Hundreds of Dimensions for Vision Architectures[C]// GOMEZ F., TUKEY J., EDWARDS R., eds. Proceedings of the 30th International Conference on Machine Learning (ICML 2013). Atlanta, GA, USA: JMLR.org, 2013: 115-123.[12] CHEN S., GOO Y.-J.J., SHEN Z.-D. A Hybrid Approach of Stepwise Regression, Logistic Regression, Support Vector Machine, and Decision Tree for Forecasting Fraudulent Financial Statements[J]. The Scientific World Journal, 2014, 11: 1-9.[13] ALAMRI M., YKHLEF M. Survey of Credit Card Anomaly and Fraud Detection Using Sampling Techniques[J]. Electronics, 2022, 11: 4003.[14]CHOI E., BISWAL S., MALIN B., DUKE J., STEWART W.F., SUN J. Generating multi-label discrete patient records using generative adversarial networks[C]// SMITH J., WANG L., eds. Proceedings of the Machine Learning for Healthcare Conference. Boston: Springer, 2017: 286-305.[15]FENG X., ZHAO K., ZENG H. Combining GANs with anomaly detection for generating high-quality minority samples[J]. Knowledge-Based Systems, 2020, 193: 105408.[16] STRELCENIA E., PRAKOONWIT S. A survey on GAN techniques for data augmentation to address imbalanced data issues in credit card fraud detection[J]. Machine Learning and Knowledge Extraction, 2023, 5(1): 304-329.[17]FAWCETT T. An introduction to ROC analysis[J]. Pattern Recognition Letters, 2006, 27(8): 861-874.[18]SUH S., LEE H., LUKOWICZ P., LEE Y.O. CEGAN: Classification Enhancement Generative Adversarial Networks for unraveling data imbalance problems[J]. Neural Networks, 2021, 133: 69-86.[19]ZAREAPOOR M., SHAMSOLOMOALI P., YANG J. Oversampling adversarial network for class-imbalanced fault diagnosis[J]. Mechanical Systems and Signal Processing, 2021, 149: 107175.[20]MULLICK Sankha Subhra S., DATTA Shounak, DAS Swagatam. Generative adversarial minority oversampling[C]//Zhou Jie, Feng Shuiwang, eds. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). Seoul, Republic of Korea: IEEE, 2019: 16951704.[21]MIRZA M., OSINDERO S. Conditional Generative Adversarial Nets[J]. Computer Science, 2014: 2672-2680.[22]BA HUNG. Improving detection of credit card fraudulent transactions using generative adversarial networks[J]. arXiv, 2019, 1907.03355.[23]SOMORJIT L., VERMA M. Variants of generative adversarial networks for credit card fraud detection[C]// KAR NIRMALYA, SAHA ASHIM, DEB SUMAN, eds. Trends in Computational Intelligence, Security and Internet of Things: Third International Conference, ICCISIoT 2020, Proceedings. Cham, Switzerland: Springer, 2020: 133-143.[24]CHEN T., GUESTRIN C. XGBoost: A scalable tree boosting system[C]// KRISHNAPURAM B., SHAH M., SMOLA A.J., AGGARWAL C.C., SHEN D., RASTOGI R., eds. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, CA, USA: ACM, 2016: 785‑794.[25] 李航. 统计学习方法[M]. 北京：清华大学出版社，2012.[26]HASTIE T., TIBSHIRANI R., FRIEDMAN J. The Elements of Statistical Learning: Data Mining, Inference, and Prediction[M]. 2nd ed. New York, NY, USA: Springer, 2009.[27]LIANG W., LUO S., ZHAO G., WU H. Predicting hard rock pillar stability using GBDT, XGBoost, and xgboost algorithms[J]. Mathematics, 2020, 8(765): 1-12.[28] 王刚., 李忠伟., 任明明., 李鹏. 机器学习- 贝叶斯和优化方法[M].2版. 北京：机械工业出版社,2021.[29]WU J., CHEN X.Y., ZHANG H., XIONG L.D., LEI H., DENG S.H. Hyperparameter Optimization for Machine Learning Models Based on Bayesian Optimization[J]. Journal of Electronic Science and Technology, 2019, 17(1): 26-40.[30]SUN X., XU W., WANG X. Bayesian Optimization for Hyperparameter Tuning in XGBoost-based Credit Scoring Model[C]// LI H., ZHANG Y., eds. Proceedings of the 2020 International Conference on Artificial Intelligence and Big Data (ICAIBD). Chengdu, China: IEEE, 2020: 393-398.[31]CHEN T., GUESTRIN C. XGBoost: A scalable tree boosting system[C]// KRISHNAPURAM B., SHAH M., SMOLA A.J., AGGARWAL C.C., SHEN D., RASTOGI R., eds. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San Francisco, CA, USA: ACM, 2016: 785-794.[32]FRIEDMAN J.H. Greedy Function Approximation: A Gradient Boosting Machine[J]. Annals of Statistics, 2001, 29(5): 1189‑1232.[33]HE H., GARCIA E.A. Learning from imbalanced data[J]. IEEE Transactions on Knowledge and Data Engineering, 2009, 21(9): 1263-1284.