EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

LLM-Explanation-Enhanced Retail Credit Default Prediction with Gradient Boosting on the UCI Default of Credit Card Clients Dataset

Hailin Zhou, Sarah Zhao

Journal of Advanced Computing Systems · 22.05.2024 · 4 (5) · 102-118

DOI: 10.69987/jacs.2024.40508 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Advanced Computing Systems, online erschienen am 22.05.2024. Verfasst von Hailin Zhou und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Retail credit scoring still depends on strong tabular learners, but operational use also requires explanations that describe behavior rather than only output a probability. This study evaluated an LLM-explanation-enhanced gradient boosting pipeline on the UCI Default of Credit Card Clients dataset. The benchmark contains 30,000 clients, 23 predictive variables, and no missing values [1], [2]. We transformed each client’s six-month bill and repayment history into a deterministic natural-language risk behavior summary that mimics an analyst-style LLM note while remaining fully reproducible. The summary encoded delinquency sequence, utilization, repayment coverage, bill trend, and explicit risk or protective tags. Structured features were modeled with XGBoost and LightGBM, summaries were modeled with TF-IDF logistic regression, and both branches were fused by fixed weighted late fusion. All numbers reported in the manuscript were empirically measured by executing the supplied code; no illustrative placeholders were retained. On a representative 70/15/15 split, XGBoost with engineered structured features reached AUROC 0.7874 and AUPRC 0.5758, while XGBoost weighted fusion reached AUROC 0.7867, AUPRC 0.5731, F1 0.5524, and accuracy 0.8044. Across five repeated stratified splits, XGBoost weighted fusion achieved the best mean AUROC (0.7943 ± 0.0111) and mean AUPRC (0.5706 ± 0.0181), and LightGBM weighted fusion reached 0.7911 ± 0.0102 AUROC and 0.5686 ± 0.0185 AUPRC. The text-only summary model remained competitive at 0.7869 ± 0.0102 AUROC, showing that the generated explanations preserved most of the discriminatory information. Ablation results showed that repayment-sequence narration, finance descriptors, and explicit risk tags all added measurable value. The findings demonstrate that explanation-oriented table-to-text serialization can improve gradient-boosted retail default prediction while simultaneously producing auditor-friendly behavioral summaries.

Zitieren

APA

Zhou, H., & Zhao, S. (2024). LLM-Explanation-Enhanced Retail Credit Default Prediction with Gradient Boosting on the UCI Default of Credit Card Clients Dataset. Journal of Advanced Computing Systems, 4(5), 102-118. https://doi.org/10.69987/jacs.2024.40508

Vancouver

Zhou H, Zhao S. LLM-Explanation-Enhanced Retail Credit Default Prediction with Gradient Boosting on the UCI Default of Credit Card Clients Dataset. Journal of Advanced Computing Systems. 2024;4(5):102-118. doi:10.69987/jacs.2024.40508

BibTeX
@article{Zhou2024,
  author = {Zhou, Hailin and Zhao, Sarah},
  title = {LLM-Explanation-Enhanced Retail Credit Default Prediction with Gradient Boosting on the UCI Default of Credit Card Clients Dataset},
  journal = {Journal of Advanced Computing Systems},
  year = {2024},
  volume = {4},
  number = {5},
  pages = {102--118},
  doi = {10.69987/jacs.2024.40508}
}
Kostenlos starten →