Han, Hanif and Mantoro, Teddy and Santoso, Handri Predicting Loan Delinquency in Installment Loans Using LightGBM for Enhanced Credit Risk Management. TEPIAN.
HANIF HAN.pdf
Download (540kB)
Abstract
Credit risk assessment is essential for financial
institutions to effectively manage loan portfolios, especially for installment loans. Predicting delinquency is challenging due to the complex interplay of borrower behavior, loan characteristics, and repayment pattern. Traditional models often fail to capture non-linear relationships in data and require significant preprocessing to address imbalanced datasets, feature scaling, and diverse data distributions, resulting in inefficiencies. This research leverages LightGBM, a gradient-boosting algorithm, to tackle these challenges and enhance delinquency prediction accuracy.
LightGBM is well-suited for handling imbalanced datasets,
employs an efficient histogram-based binning method, and
reduces preprocessing needs by managing features of varying
scales inherently. The model combines borrower attributes,
loan-specific details, and engineered payment history
features, with optimization achieved through feature
engineering and hyperparameter tuning. The evaluation,
conducted on real-world loan data, demonstrates that the
LightGBM model significantly outperforms traditional
approaches in precision, recall, and AUC metrics. The study
offers a scalable and practical tool for credit risk management while emphasizing the potential for enhanced predictive performance. Future work could incorporate macroeconomic factors and assess real-time deployment to further expand the model’s applicability.
Keywords: Credit Risk Assessment, Loan Delinquency
Prediction, Installment Loans, LightGBM, Feature Engineering, Machine Learning in Finance
| Item Type: | Article |
|---|---|
| Subjects: | -|- SUBJEK PRADITA -|- > Fakultas Sains dan Teknologi > Magister Teknologi Informasi |
| Divisions: | Fakultas Sains dan Teknologi > Magister Teknologi Informasi |
| Depositing User: | Pradita Librarian |
| Date Deposited: | 23 Oct 2026 03:18 |
| Last Modified: | 23 Oct 2026 03:18 |
| URI: | https://repository.pradita.ac.id/id/eprint/797 |
