Gradient Descent Decision Tree Algorithm and Nonlinear Programming for Credit Risk Assessment and Credit Strategy
Downloads
This research aimed to develop a scientific and accurate credit risk assessment model for small and medium-sized enterprises (SMEs) to support banks in credit decision-making. An improved decision tree model is proposed by integrating regularization to control complexity and employing an ensemble learning approach to enhance prediction accuracy. Multiple weak classifiers are iteratively refined using gradient descent optimization to form a robust, strong classifier. The model is trained through supervised learning, with the default probability of SMEs as the objective function, enabling a quantitative assessment of credit risk. The findings show that the proposed gradient descent decision tree algorithm achieves an AUC of 0.99 under 70% and 80% training set ratios, outperforming Adaptive Boosting (AUC = 0.97), Random Forest (AUC = 0.91), and traditional decision trees (AUC = 0.82). To further optimize bank loan strategies, this paper constructs a nonlinear multi-objective programming model that maximizes expected loan returns while considering risk constraints. The proposed approach not only improves credit risk prediction but also assists banks in formulating optimal lending strategies. This study advances credit risk modelling by integrating regularization and ensemble learning, offering a novel and practical solution for SME credit assessment.
Downloads
[1] Djebali, N., & Zaghdoudi, K. (2020). Threshold effects of liquidity risk and credit risk on bank stability in the MENA region. Journal of Policy Modeling, 42(5), 1049–1063. doi:10.1016/j.jpolmod.2020.01.013.
[2] Dodd, O., Kalimipalli, M., & Chan, W. (2021). Evaluating corporate credit risks in emerging markets. International Review of Financial Analysis, 73, 101–13. doi:10.1016/j.irfa.2020.101610.
[3] He, S. S., Hou, W. H., Chen, Z. Y., Liu, H., Wang, J. Q., & Cheng, P. F. (2025). Early warning model based on support vector machine ensemble algorithm. Journal of the Operational Research Society, 76(3), 411-425. doi:10.1080/01605682.2024.2360111.
[4] Kaur, S., Singh, A., & Aggarwal, A. (2024). Mean-Variance optimal portfolio selection integrated with Support Vector and Fuzzy Support Vector Machines. Journal of Fuzzy Extension and Applications, 5(3), 434–468. doi:10.22105/jfea.2024.453926.1453.
[5] Doumpos, M., & Zopounidis, C. (2007). Model combination for credit risk assessment: A stacked generalization approach. Annals of Operations Research, 151(1), 289–306. doi:10.1007/s10479-006-0120-x.
[6] Tandel, V., Gandhi, S., & Tabarrok, A. (2023). Building networks: Investigating the quid pro quo between local politicians & developers. Journal of Development Economics, 164, 103138. doi:10.1016/j.jdeveco.2023.103138.
[7] Quan, J., & Sun, X. (2024). Credit risk assessment using the factorization machine model with feature interactions. Humanities and Social Sciences Communications, 11(1), 1–10. doi:10.1057/s41599-024-02700-7.
[8] Yu, T., Huang, W., Tang, X., & Zheng, D. (2025). A hybrid unsupervised machine learning model with spectral clustering and semi-supervised support vector machine for credit risk assessment. PLoS ONE, 20(1), 316557. doi:10.1371/journal.pone.0316557.
[9] Carling, K., Jacobson, T., Lindé, J., & Roszbach, K. (2007). Corporate credit risk modeling and the macroeconomy. Journal of Banking and Finance, 31(3), 845–868. doi:10.1016/j.jbankfin.2006.06.012.
[10] Zhu, S. (2025). Research on Credit Rating of Commercial Banks Based on Support Vector Machine—Data from China’s Listed Commercial Banks as Evidence. Finance, 15(02), 423–429. doi:10.12677/fin.2025.152044.
[11] Jarrow, R. A., & Turnbull, S. M. (1995). Pricing Derivatives on Financial Securities Subject to Credit Risk. The Journal of Finance, 50(1), 53–85. doi:10.1111/j.1540-6261.1995.tb05167.x.
[12] Giudici, P., Hadji-Misheva, B., & Spelta, A. (2020). Network based credit risk models. Quality Engineering, 32(2), 199–211. doi:10.1080/08982112.2019.1655159.
[13] Guo, Y., Zhou, W., Luo, C., Liu, C., & Xiong, H. (2016). Instance-based credit risk assessment for investment decisions in P2P lending. European Journal of Operational Research, 249(2), 417–426. doi:10.1016/j.ejor.2015.05.050.
[14] Bulut, C., & Arslan, E. (2024). Comparison of the impact of dimensionality reduction and data splitting on classification performance in credit risk assessment. Artificial Intelligence Review, 57(9), 252 10 1007 10462–024–10904–1. doi:10.1007/s10462-024-10904-1.
[15] Kubo, H., & Sakai, Y. (2011). On long-term credit risk assessment and rating: Towards a new set of models. Journal of Risk Research, 14(9), 1127–1141. doi:10.1080/13669877.2011.571793.
[16] Shen, C., & Wu, J. (2025). Research on credit risk of listed companies: a hybrid model based on TCN and DilateFormer. Scientific Reports, 15(1), 2599. doi:10.1038/s41598-025-86371-7.
[17] Tarigan, P. A., & Manurung, A. H. (2024). Determination of Bank RAROC using Internal and External Factors of the Bank and Size as a Moderating Variable. Ekonomis: Journal of Economics and Business, 8(2), 1611. doi:10.33087/ekonomis.v8i2.2063.
[18] Lopez, J. A., & Saidenberg, M. R. (2000). Evaluating credit risk models. Journal of Banking and Finance, 24(1–2), 151–165. doi:10.1016/S0378-4266(99)00055-2.
[19] Oreski, S., & Oreski, G. (2014). Genetic algorithm-based heuristic for feature selection in credit risk assessment. Expert Systems with Applications, 41(4), 2052–2064. doi:10.1016/j.eswa.2013.09.004.
[20] Luo, S., Kong, X., & Nie, T. (2016). Spline based survival model for credit risk modeling. European Journal of Operational Research, 253(3), 869–879. doi:10.1016/j.ejor.2016.02.050.
[21] Liu, J., Liu, J., Wu, C., & Wang, S. (2024). Enhancing credit risk prediction based on ensemble tree-based feature transformation and logistic regression. Journal of Forecasting, 43(2), 429–455. doi:10.1002/for.3040.
[22] Stefania, G. C., Claudia, G. D., Guillermo, A. M. L., Gustavo, G., Dario, R. H. J., Alfonso, M. T. F., & Tatiana, A. N. S. (2023). Credit Risk Scoring Model Based on the Discriminant Analysis Technique. Procedia Computer Science, 220, 928–933. doi:10.1016/j.procs.2023.03.127.
[23] Wang, Z., Zhang, X., Zhang, Z. K., & Sheng, D. (2022). Credit portfolio optimization: A multi-objective genetic algorithm approach. Borsa Istanbul Review, 22(1), 69–76. doi:10.1016/j.bir.2021.01.004.
[24] Zhou, X., Jiang, W., & Shi, Y. (2010). Credit risk evaluation by using nearest subspace method. Procedia Computer Science, 1(1), 2449–2455. doi:10.1016/j.procs.2010.04.276.
[25] Kotelnikova, Y., Clark, L. A., Vernon, P. A., & Hayden, E. P. (2014). Development and Validation of the Schedule for Nonadaptive and Adaptive Personality Brief Self-Description Rating Form (SNAP-BSRF). Assessment, 22(1), 3–16. doi:10.1177/1073191114534959.
[26] Kallmuenzer, A., Mikhaylov, A., Chelaru, M., & Czakon, W. (2025). Adoption and performance outcome of digitalization in small and medium-sized enterprises. Review of Managerial Science, 19(7), 2011–2038. doi:10.1007/s11846-024-00744-2.
[27] Liang, Y., Zhou, B., & Zhao, S. (2024). Risking or de-risking? The effect of banking competition on large state-owned banks and small and medium-sized enterprise lending: Evidence from China. International Review of Financial Analysis, 94, 103258. doi:10.1016/j.irfa.2024.103258.
[28] Judijanto, L., Hairuddin, S. H., Subhan, S., & Sipayung, B. (2024). Analysis of the Effect of Risk Management and Compliance Practices on Financial Performance and Corporate Reputation in the Financial Industry in Indonesia. The Es Accounting and Finance, 2(03), 177–191. doi:10.58812/esaf.v2i03.293.
[29] Carè, R., Fatima, R., & Lèvy, N. (2024). Assessing the evolution of banking reputation literature: a bibliometric analysis. International Journal of Bank Marketing, 42(5), 1059–1091. doi:10.1108/IJBM-07-2023-0417.
[30] Kim, S., & Kim, E. (2024). Population decline in small and medium-sized cities and spatial economic patterns: spatial probit model of South Korea. Applied Economics Letters, 1–6. doi:10.1080/13504851.2024.2363303.
[31] De la Llave, M., & López, F. A. (2024). Searching for correct specification in spatial probit models. Classical approaches versus Gradient Boosting algorithm. Spatial Statistics, 61, 100815. doi:10.1016/j.spasta.2024.100815.
[32] Campanella, F., Ferri, L., Serino, L., & Zampella, A. (2025). Exploring the link between sustainable performance and credit access: the moderating role of intellectual capital. Journal of Intellectual Capital, 26(1), 205–228. doi:10.1108/JIC-06-2024-0191.
[33] Hou, L., Lu, K., & Bi, G. (2024). Predicting the credit risk of small and medium-sized enterprises in supply chain finance using machine learning algorithms. Managerial and Decision Economics, 45(4), 2393–2414. doi:10.1002/mde.4130.
[34] Li, S. (2024). Leveraging Big Data for SME Credit Risk Assessment: A Novel BP-KMV and GARCH Integration. Journal of the Knowledge Economy, 1–29. doi:10.1007/s13132-024-01995-w.
[35] Hou, L., Bi, G., & Guo, Q. (2025). An improved sparrow search algorithm optimized LightGBM approach for credit risk prediction of SMEs in supply chain finance. Journal of Computational and Applied Mathematics, 454, 116197. doi:10.1016/j.cam.2024.116197.
[36] Fakir, Y., Azalmad, M., & Elaychi, R. (2020). Study of The ID3 and C4.5 Learning Algorithms. Journal of Medical Informatics and Decision Making, 1(2), 29–43. doi:10.14302/issn.2641-5526.jmid-20-3302.
[37] Chanmee, S., & Kesorn, K. (2023). Semantic decision Trees: A new learning system for the ID3-Based algorithm using a knowledge base. Advanced Engineering Informatics, 58, 102156. doi:10.1016/j.aei.2023.102156.
[38] Mei, L. (2025). Application and Optimization of Decision Tree ID3 Algorithm in Innovation and Entrepreneurship Analysis Model. International Journal of High Speed Electronics and Systems, 34(2). doi:10.1142/S0129156424400913.
[39] Cirgon, B. Z. (2025). Comparison of C4.5 and C5.0 Methods for Classification of Bad Credit and Good Credit. Formosa Journal of Science and Technology, 4(2), 789–808. doi:10.55927/fjst.v4i2.16.
[40] Ileberi, E., Sun, Y., & Wang, Z. (2021). Performance Evaluation of Machine Learning Methods for Credit Card Fraud Detection Using SMOTE and AdaBoost. IEEE Access, 9, 165286–165294. doi:10.1109/ACCESS.2021.3134330.
[41] Wang, S., Liu, W., Yang, S., & Huang, H. (2024). An optimized AdaBoost algorithm with atherosclerosis diagnostic applications: adaptive weight-adjustable boosting. Journal of Supercomputing, 80(9), 13187–13216. doi:10.1007/s11227-024-05951-y.
[42] Asbai, N., Bounazou, H., & Zitouni, S. (2025). A novel approach to deriving adaboost classifier weights using squared loss function for overlapping speech detection. Multimedia Tools and Applications, 1-28. doi:10.1007/s11042-025-20718-0.
- This work (including HTML and PDF Files) is licensed under a Creative Commons Attribution 4.0 International License.
