Research on Dynamic Credit Risk Prediction in Commercial Banks Based on a Hybrid Scorecard-LSTM-Attention Model
DOI:
https://doi.org/10.54097/h8p4w333Keywords:
Credit Risk, Credit Scorecard, Attention Mechanism, LSTM, Neural NetworkAbstract
With the increasing fluctuations in the global economy and the development of digital finance in China, commercial banks are facing two major challenges: the declining quality of credit assets and the growth of bad loans. Traditional credit scoring models based on Logistic Regression are easy to explain, but they struggle to handle complex data that is high-dimensional or changes over time. To improve the accuracy of credit risk prediction, this study develops a hybrid model called Scorecard-LSTM. This model combines traditional credit scoring methods with LSTM networks, also added an Attention Mechanism to help the model focus more on key customer behaviors. The study uses data from a commercial bank in Taiwan, including basic customer information, financial status, and repayment history. By using methods like WOE encoding, feature binning, and joint training, the model connects the clear logic of traditional scorecards with the powerful learning ability of deep learning. The experimental results show that this hybrid model is much better than traditional models in both accuracy and stability. With an AUC value of 0.7741, the model performs very well in identifying bad loans and providing early warnings for potential defaults.
Downloads
References
[1] Peussa, A. (2016). Credit risk scorecard estimation by logistic regression [Unpublished manuscript]. University of Helsinki, Faculty of Science, Department of Mathematics and Statistics.
[2] Ma, H. (2023). Risk management of commercial banks under the background of big data. Market Modernization, (5), 95–97.
[3] Zhu, X., Cheng, D., Zhang, Z., Lin, S., & Dai, J. (2019). An empirical study of spatial attention mechanisms in deep networks. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 6688–6697). IEEE. https://doi.org/10.1109/ICCV.2019.00678
[4] Dang, Y., Miao, Z. Q., Zhang, T., & Feng, D. F. (2022). Application progress of big data methods in systemic financial risk monitoring and early warning. Journal of Financial Development Research, 2(3).
[5] Liu, P. (2025). Application of LSTM-Attention hybrid model in financial risk prediction. Information & Computer, 37(13), 43–45.
[6] Feng, Y. C. (2025). Research on the application and optimization of big data-based risk management framework in financial institutions. In Proceedings of the Academic Seminar on Smart Building and Intelligent Economic Construction (I) (pp. 248–251).
[7] Cui, X. N. (2025, February 6). Application of machine learning algorithms in financial market risk analysis and prediction. Henan Economic News, p. 9.
[8] Xu, E. X. (2024). Research on personal credit risk based on random forest model. Audit and Wealth Management, (9), 55–58.
[9] Zhang, Z. T. (2024). Application research of gradient boosting method in credit risk assessment. Automation Application, 65(13), 26–31.
[10] Zhang, L., et al. (2023). Graph neural networks for systemic risk assessment in credit guarantee networks. Nature Computational Science, 3(1), 78–92. https://doi.org/10.1038/s43588-022-00386-7
[11] Kumar, A., et al. (2022). Dynamic graph risk: Real-time credit risk prediction using evolving social networks. In 2022 IEEE International Conference on Big Data (pp. 345–358). IEEE. https://doi.org/10.1109/BigData55660.2022.9975643
[12] Xu, H. Y. (2024). Research on credit risk prediction based on machine learning [Master’s thesis]. Jilin University of Finance and Economics.
[13] Zhang, J. L., Guo, S. Y., Ren, C. P., & Ma, Q. (2024). Research on personal credit scoring card model based on logistic regression. Modern Information Technology, 8, 12–16.
[14] Dong, G., Lai, K. K., & Yen, J. (2010). Credit scorecard based on logistic regression with random coefficients. Procedia Computer Science, 1(1), 2463–2468. https://doi.org/10.1016/j.procs.2010.04.273
[15] Yu, Y., Si, X., Hu, C., & Zhang, J. (2019). A review of recurrent neural networks: LSTM cells and network architectures. Neural Computation, 31(7), 1235–1270. https://doi.org/10.1162/neco_a_01199
[16] Staudemeyer, R. C., & Morris, E. R. (2019). Understanding LSTM—A tutorial into long short-term memory recurrent neural networks [Preprint]. arXiv. https://arxiv.org/abs/1909.09586
[17] Niu, Z., Zhong, G., & Yu, H. (2021). A review on the attention mechanism of deep learning. Neurocomputing, 452, 48–62. https://doi.org/10.1016/j.neucom.2021.04.069
[18] Brauwers, G., & Frasincar, F. (2023). A general survey on attention mechanisms in deep learning. IEEE Transactions on Knowledge and Data Engineering, 35(4), 3279–3298. https://doi.org/10.1109/TKDE.2021.3067332
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
