{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,4]],"date-time":"2026-07-04T02:23:13Z","timestamp":1783131793658,"version":"3.54.6"},"reference-count":25,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T00:00:00Z","timestamp":1773878400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Systems"],"abstract":"<jats:p>In the context of the rapid advancement of financial technology, the issue of credit card default has become increasingly salient, emerging as one of the crucial risks that financial institutions are eagerly addressing. Traditional credit card default risk prediction models predominantly rely on statistical correlations for feature selection. This approach not only makes it challenging to uncover the genuine causal relationships between variables but also leads to limitations in prediction accuracy and interpretability. To overcome these limitations, this paper presents a novel credit card default risk prediction model that integrates causal feature screening, interaction feature construction, and interpretability enhancement. Initially, by leveraging the information value (IV) and eXtreme gradient boosting (XGBoost), we perform initial feature dimensionality reduction. Subsequently, we introduce the Peter Clark algorithm (PC) augmented with perturbation enhancement and bootstrap sampling to identify a stable set of causal features. Building on this foundation, we proceed to construct higher-order interaction features to bolster the model\u2019s nonlinear modeling capacity. These causal features and their interaction counterparts are then fed into a variety of mainstream machine learning models for training and evaluation purposes. Furthermore, on the basis of the causal feature set identified via the PC algorithm, we construct a causal path diagram. We also incorporate the causal forest double machine learning (causal forest DML) method to estimate the causal effects of features. Additionally, we design a counterfactual explanation mechanism to aid in analyzing the direction and magnitude of the impact of variable interventions on default probability. Empirical tests conducted using four typical credit datasets reveal the following findings: (1) the introduction of causal features generally enhances the model\u2019s performance in terms of the F1 score, area under the curve (AUC), and geometric mean (G-mean). This improvement is especially pronounced in models that are highly reliant on feature quality, such as logistic regression (LR). (2) Causal features offer significant advantages in terms of model interpretability, stability, and compliance, thereby presenting a new research paradigm for credit risk prevention and control in high-risk financial scenarios.<\/jats:p>","DOI":"10.3390\/systems14030327","type":"journal-article","created":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T15:38:01Z","timestamp":1773934681000},"page":"327","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["An Interpretable Credit Default Risk Prediction Framework Integrating Causal Feature Selection and Double Machine Learning"],"prefix":"10.3390","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6061-2517","authenticated-orcid":false,"given":"Tinggui","family":"Chen","sequence":"first","affiliation":[{"name":"School of Statistics and Data Science, Zhejiang Gongshang University, Hangzhou 310018, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rui","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Statistics and Data Science, Zhejiang Gongshang University, Hangzhou 310018, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-1647-8217","authenticated-orcid":false,"given":"Jian","family":"Hou","sequence":"additional","affiliation":[{"name":"School of Economics and Management, Shanghai Maritime University, Shanghai 201306, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,3,19]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"150","DOI":"10.1016\/j.asoc.2016.02.025","article-title":"Technology credit scoring model with fuzzy logistic regression","volume":"43","author":"Sohn","year":"2016","journal-title":"Appl. Soft Comput."},{"key":"ref_2","first-page":"2139","article-title":"Establishment of business loan default prediction model by integrating survival analysis with logistic regression","volume":"31","author":"Chang","year":"2024","journal-title":"Sci. Iran."},{"key":"ref_3","first-page":"817","article-title":"Credit Risk Prediction of Enterprises Based on Optimal Index Combination","volume":"30","author":"Zhou","year":"2021","journal-title":"J. Syst. Manag."},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Zhang, X., Zhang, T., Hou, L., Liu, X., Guo, Z., Tian, Y., and Liu, Y. (2025). Data-Driven Loan Default Prediction: A Machine Learning Approach for Enhancing Business Process Management. Systems, 13.","DOI":"10.3390\/systems13070581"},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Dichev, A., Zarkova, S., and Angelov, P. (2025). Machine Learning as a Tool for Assessment and Management of Fraud Risk in Banking Transactions. J. Risk Financ. Manag., 18.","DOI":"10.3390\/jrfm18030130"},{"key":"ref_6","first-page":"767","article-title":"Construction of a financial default risk prediction model based on the LightGBM algorithm","volume":"31","author":"Gao","year":"2022","journal-title":"J. Intell. Syst."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"324","DOI":"10.54254\/2755-2721\/49\/20241698","article-title":"Credit Default Prediction Model Based on Transformer Encoder and Residual Network","volume":"44","author":"Zhang","year":"2024","journal-title":"Comput. Appl."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"109597","DOI":"10.1016\/j.engappai.2024.109597","article-title":"WIGNN: An adaptive graph-structured reasoning model for credit default prediction","volume":"139","author":"Yan","year":"2025","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref_9","first-page":"481","article-title":"Research on Early Warning of Default Risk of Listed Companies Based on Stacking Method","volume":"45","author":"Chi","year":"2025","journal-title":"Syst. Eng.-Theory Pract."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"231","DOI":"10.1016\/j.aej.2024.11.054","article-title":"A probabilistic approach driven credit card anomaly detection with CBLOF and isolation forest models","volume":"114","author":"Chugh","year":"2025","journal-title":"Alex. Eng. J."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"12750","DOI":"10.1109\/ACCESS.2021.3052054","article-title":"Bond Default Prediction Based on Deep Learning and Knowledge Graph Technology","volume":"9","author":"Ma","year":"2021","journal-title":"IEEE Access"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"8607","DOI":"10.1007\/s00521-024-09546-9","article-title":"CFTNet: A robust credit card fraud detection model enhanced by counterfactual data augmentation","volume":"36","author":"Kong","year":"2024","journal-title":"Neural Comput. Appl."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"586","DOI":"10.1016\/j.ejor.2024.09.025","article-title":"Attention-based dynamic multilayer graph neural networks for loan default prediction","volume":"321","author":"Zandi","year":"2025","journal-title":"Eur. J. Oper. Res."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"122825","DOI":"10.1016\/j.techfore.2023.122825","article-title":"Interpretable high-stakes decision support system for credit default forecasting","volume":"196","author":"Sun","year":"2023","journal-title":"Technol. Forecast. Soc. Change"},{"key":"ref_15","unstructured":"Kagaba, D., Busingye, C., Nansamba, B., Jingo, D., and Marvin, G. (2024). Explainable deep ensemble learning for improved credit default prediction. Proceedings of the 2024 Sixteenth International Conference on Contemporary Computing, Association for Computing Machinery."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Arias-Serna, M.A., Quiza-Montealegre, J.J., M\u00f3ntes-G\u00f3mez, L.F., Uribe Clavijo, L., and Orozco-Duque, A.F. (2025). Explainable Machine Learning Models for Credit Rating in Colombian Solidarity Sector Entities. J. Risk Financ. Manag., 18.","DOI":"10.20944\/preprints202507.0756.v1"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Pearle, J. (2009). Causality: Models, Reasoning, and Inference, Cambridge University Press. [2nd ed.].","DOI":"10.1017\/CBO9780511803161"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"541","DOI":"10.1016\/j.bir.2025.02.006","article-title":"Revisiting the macroeconomic determinants of non-performing loans with a deep learning technique with causal inference: Evidence from T\u00fcrkiye","volume":"25","author":"Bakir","year":"2025","journal-title":"Borsa Istanb. Rev."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"106985","DOI":"10.1016\/j.econmod.2024.106985","article-title":"Forecasting China bond default with severe class-imbalanced data: A simple learning model with causal inference","volume":"144","author":"Peng","year":"2025","journal-title":"Econ. Model."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"19","DOI":"10.3934\/DSFE.2025002","article-title":"Prediction of bank credit customers churn based on machine learning and interpretability analysis","volume":"5","author":"Li","year":"2025","journal-title":"Data Sci. Financ. Econ."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Noh, M.J., and Kim, Y.S. (2025). Diabetes prediction through linkage of causal discovery and inference model with machine learning models. Biomedicines, 13.","DOI":"10.3390\/biomedicines13010124"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"124665","DOI":"10.1016\/j.jenvman.2025.124665","article-title":"An efficient catalyst screening strategy combining machine learning and causal inference","volume":"377","author":"Song","year":"2025","journal-title":"J. Environ. Manag."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"12","DOI":"10.1186\/s10020-023-00603-y","article-title":"Gene targeting in amyotrophic lateral sclerosis using causality-based feature selection and machine learning","volume":"29","author":"Founta","year":"2023","journal-title":"Mol. Med."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"e70498","DOI":"10.1002\/cpe.70498","article-title":"Credit Risk Identification Algorithm Based on BaggingFCBF-TCN","volume":"38","author":"Chen","year":"2026","journal-title":"Concurr. Comput. Pract. Exp."},{"key":"ref_25","first-page":"1","article-title":"A credit card fraud detection model based on a stacked temporospatial graph attention residual network","volume":"22","author":"Chen","year":"2026","journal-title":"J. Credit Risk"}],"container-title":["Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2079-8954\/14\/3\/327\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,21]],"date-time":"2026-03-21T05:17:07Z","timestamp":1774070227000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2079-8954\/14\/3\/327"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,19]]},"references-count":25,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3]]}},"alternative-id":["systems14030327"],"URL":"https:\/\/doi.org\/10.3390\/systems14030327","relation":{},"ISSN":["2079-8954"],"issn-type":[{"value":"2079-8954","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,19]]}}}