Research on Credit Card Fraud Risk Identification Based on Integrated Logistic Regression Model

Mo Yu; Rongpeng Yan

doi:10.54097/eq3njd96

Authors

Mo Yu
Rongpeng Yan

DOI:

https://doi.org/10.54097/eq3njd96

Keywords:

Credit card fraud risk assessment, resample principle, Logistic Regression, Ensemble Learning.

Abstract

Effectively identifying credit card fraud using machine learning methods is a significant issue in the financial sector. In this context, machine learning models encounter challenges such as the imbalanced distribution of sample data labels and the high dimensionality of customer feature sets. Addressing these two critical factors, this paper develops an enhanced method for the logistic regression model. This approach not only balances the sample label distribution through resampling but also mitigates the estimation issues arising from the curse of dimensionality. Furthermore, the proposed method addresses the coverage issue of the entire feature set. It solves that resampling can only partially address the curse of dimensionality problem and employs L1 regularization for each logistic regression submodel to further alleviate this issue. Results from simulation experiments and real-world data analysis demonstrate that the proposed method is competitive with logistic regression and several classical classification techniques. This method is not only effective in resolving credit card fraud risks but also has the potential to be extended to other domains.

Downloads

Download data is not yet available.

References

[1] FOROUGH J,MOMTAZI S.Ensemble of deep sequential models for credit card fraud detection[J].Applied Soft Computing,2021,99:106883.

[2] Mei Y ,K. M L ,Yingchi Q , et al.Deep neural networks with L1 and L2 regularization for high dimensional corporate credit risk prediction[J].Expert Systems With Applications,2023,213(PA):

[3] Agarwal A ,Rana A,Verma N, et al.Enhancement of classification techniques using principal component analysis and class imbalance handling methods in credit card defaulter detection[J].International Journal of Forensic Engineering,2021,5(1):1-18.

[4] Wang H ,Hancock J ,Khoshgoftaar M T .Improving Credit Card Fraud Detection with Data Reduction Approaches[J].International Journal of Reliability, Quality and Safety Engineering,2024,31(04):

[5] Mosa T D ,Sorour E S ,Abohany A A , et al.CCFD: Efficient Credit Card Fraud Detection Using Meta-Heuristic Techniques and Machine Learning Algorithms[J].Mathematics,2024,12(14):2250-2250.

[6] Tang Y ,Liang Y .Credit card fraud detection based on federated graph learning[J].Expert Systems With Applications,2024,256124979-124979.

[7] Zhang Junli, Guo Shuangyan, Ren Cuiping, et al. Study on the personal credit score card model based on logistic regression [J]. Modern Information Technology, 2024,8(05):12-16.DOI:10.19850/j.cnki.2096-4706.2024.05.003.

[8] Du Kang Le. Stochastic optimization algorithm [D] with the L_1 regularization problem. Zhejiang Normal University, 2023.DOI:10.27464/d.cnki.gzsfu. 2023.001931.

[9] Jiang Hongxun, Jiang Junyi, Liang Xun. Review of machine learning-based research on fraud detection of credit card transactions [J]. Computer Engineering and Application, 2023,59 (21): 1-25.

[10] Zhang Junli, Guo Shuangyan, Ren Cuiping, et al. Study on the personal credit score card model based on logistic regression [J]. Modern Information Technology, 2024,8(05):12-16.DOI:10.19850/j.cnki.2096-4706.2024.05.003.

[11] Chen Shou, Yu Xiuyun, Qiu Yongqin, et al. Credit score model based on a semi-supervised SVM [J]. Management Science in China, 2024,32(03):1-8.DOI:10.16381/j.cnki.issn1003-207x.2021.2434.

[12] Ju Chunhua, Chen Guanyu, Bao Fuguang. Consumer finance risk detection model based on kNN-Note-LSTM —— Take credit card fraud detection as an example [J]. Systems Science and Mathematics, 2021,41 (02): 481-498.

[13] DOUZAS G, BACAO F, LAST Improving imbalanced learning through a heuristic oversampling method based on k-means and SMOTE [J]. Information sciences,2018,465:1-20.