Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
October 2, 2025Open Access

Explainable Artificial Intelligence Credit Risk Assessment using Machine Learning

View Full Paper
Ask AI
Bookmark
Share

Authors

SSShreya SharmaNational Institute of Science Communication and Information ResourcesHPHarsh PathakPurdue University West Lafayette

Discussion

Loading...

Member takes

Implication

This analysis demonstrates credit risk assessment using machine learning models, suggesting transparency in AI-driven decisions.

Key Points

  • LightGBM provides the highest accuracy in assessing credit risks, optimizing the balance between approvals and defaults.
  • Performance metrics like ROC-AUC, precision, recall, and F1-score were evaluated to ensure model effectiveness and reliability.
  • Custom imputation and standardization were incorporated in data preprocessing to enhance model predictions while managing class imbalance.
  • Applicant-specific reports generated by the system enhance decision-making transparency, utilizing explainable AI techniques like SHAP and LIME.

Cite This Study

Sharma et al. (2025) studied this question.

synapsesocial.com/papers/68de84bf5b556a9128e1be49https://doi.org/10.48550/arxiv.2506.19383
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Explainable Artificial Intelligence for Credit Risk Assessment: Balancing Transparency and Predictive Performance2025 · 3 citations
  2. 2Credit Risk Modeling with Explainable AI: Predictive Approaches for Loan Default Reduction in Financial Institutions2021 · 17 citations
  3. 3Credit Risk Analysis using Explainable Artificial Intelligence2024 · 2 citations
  4. 4Predictive Modelling of Credit Default Risk Using Machine Learning and Ensemble Techniques2026 · 3 citations
  5. 5Artificial Intelligence In Financial Decision Systems: Transforming Risk Assessment And Investment Practices In The Era Of Digital Scientific Culture2025