Evaluasi Komprehensif Model Particle Swarm Optimization–Support Vector Machine untuk Prediksi Risiko Kredit Konsumen

Dedi Saputra; Eva  Meilinda; Fajar  Akbar; Juniato  Sidauruk

doi:10.59395/jitp.v6i1.192

Authors

Dedi Saputra Universitas Bina Sarana Informatika
Eva Meilinda Universitas Bina Sarana Informatika
Fajar Akbar Informatika, Universitas Nusa Mandiri
Juniato Sidauruk Universitas Bina Sarana Informatika

DOI:

https://doi.org/10.59395/jitp.v6i1.192

Keywords:

credit scoring, risiko kredit, particle swarm optimization, support vector machine, explainable artificial intelligence, SHAP, machine learning

Abstract

Penilaian risiko kredit merupakan komponen penting dalam industri keuangan karena berpengaruh langsung terhadap kualitas pengambilan keputusan dan tingkat risiko pembiayaan. Berbagai pendekatan machine learning telah dikembangkan untuk meningkatkan akurasi prediksi risiko kredit, termasuk integrasi Particle Swarm Optimization dan Support Vector Machine (PSO-SVM). Meskipun pendekatan tersebut telah menunjukkan kinerja yang menjanjikan, sebagian besar penelitian sebelumnya masih berfokus pada peningkatan akurasi klasifikasi, sementara aspek stabilitas model, interpretabilitas hasil, dan perbandingan terhadap algoritma modern masih relatif terbatas. Penelitian ini bertujuan mengevaluasi kembali efektivitas PSO-SVM dalam prediksi risiko kredit melalui kerangka evaluasi yang lebih komprehensif. Particle Swarm Optimization digunakan untuk melakukan optimasi parameter dan seleksi fitur, sedangkan Support Vector Machine digunakan sebagai model klasifikasi utama. Untuk memperoleh gambaran kinerja yang lebih menyeluruh, model yang diusulkan dibandingkan dengan beberapa algoritma pembelajaran mesin yang umum digunakan dalam credit scoring, yaitu Logistic Regression, Random Forest, dan Extreme Gradient Boosting (XGBoost). Evaluasi dilakukan menggunakan Stratified Repeated 10-Fold Cross Validation dengan metrik Accuracy, Precision, Recall, F1-Score, dan Area Under the Receiver Operating Characteristic Curve (AUC-ROC). Selain itu, pendekatan Explainable Artificial Intelligence melalui SHapley Additive exPlanations (SHAP) diterapkan untuk mengidentifikasi kontribusi masing-masing fitur terhadap keputusan model. Hasil penelitian diharapkan memberikan pemahaman yang lebih mendalam mengenai kemampuan prediktif, stabilitas, dan interpretabilitas model PSO-SVM pada permasalahan risiko kredit. Kontribusi utama penelitian ini terletak pada penyediaan kerangka evaluasi yang lebih komprehensif dibandingkan penelitian terdahulu, sehingga dapat memberikan dasar empiris yang lebih kuat dalam pengembangan sistem pendukung keputusan kredit berbasis kecerdasan buatan.

Downloads

Download data is not yet available.

References

Ayari, H., Ben Said, F., & Boujelbene, Y. (2025). Machine learning powered financial credit scoring: A systematic literature review. Artificial Intelligence Review.

Bahadori, S. (2026). A comprehensive literature review on AI-based credit scoring. Expert Systems with Applications.

Bahlool, R., Hewahi, N., & Elmedany, W. (2026). Performance, fairness, and explainability in AI-based credit scoring: A systematic literature review. Journal of Risk and Financial Management, 19(2), 104.

Chacon, D. U. (2026). An explainable machine learning model for consumer credit scoring using non-traditional data. Expert Systems with Applications.

Engan, A. (2026). A review of credit risk assessment in banking. Journal of Financial Regulation and Compliance.

Hadji Misheva, B., Osterrieder, J., Hirsa, A., Kulkarni, O., & Lin, S. F. (2021). Explainable AI in Credit Risk Management. ArXiv.

Haque Mukit, M. M., Rahman, M. S., & Islam, M. T. (2026). Machine Learning & Artificial Intelligence Powered Credit Scoring Systems: A Review of Contemporary Approaches. Risks, 14(1), 12.

Japinye, A. O., & Adedugbe, A. A. (2026). Explainable AI for Credit Scoring with SHAP-Calibrated Ensembles: A Multi-Market Evaluation on Public Lending Data. SSR Journal of Artificial Intelligence.

John, S., Albert, C., & Ebenezer, E. (2025). xplainable AI (XAI) Using SHAP and LIME for Financial Fraud Detection and Credit Scoring. Proceedings of the International Conference on Advanced Computing Technologies.

Kakkar, S. (2025). Explainable AI Models for Credit Risk Scoring in Banking. International Journal of Financial Data Science.

Kumar, D. (2025). Explainable Machine Learning Models for Credit Risk Classification. SSRN Working Paper.

Lin, L., et al. (2025). SHAP Stability in Credit Risk Management: A Case Study. Risks, 13(12), 238.

Machado, M. R., & Al, E. (2025). An analytical approach to credit risk assessment using machine learning. Results in Engineering.

Meilinda, E., Saputra, D., & Sidauruk, J. (2025). Implementasi model prototype pada sistem informasi pendataan pompa ukur bbm pada spbu. 8, 599–608. https://doi.org/10.37600/tekinkom.v8i2.2518

Mestiri, S., Ben Jabeur, S., & Hamdi, B. (2024). Credit scoring using machine learning and deep learning-based models. Data Science in Finance and Economics, 4(2), 236–248.

Nallakaruppan, M. K., Kumar, R., & Singh, A. (2024). An explainable AI framework for credit evaluation and financial decision support. Applied Soft Computing, 150.

Paz, Á., et al. (2025). Machine Learning and Metaheuristics Approach for Credit Risk Prediction: A Systematic Literature Review. Biomimetics.

Pratiwi, C. (2018). Pelaksanaan Pelayanan Prima Customer Service Pada Pt Remala Abadi Bekas. Journal of Chemical Information and Modeling, 53(9), 8–24.

Rafi, M. A. (2024). Explainable AI for Credit Risk Assessment: A Data-Driven Approach. Journal of Economics, Finance and Administrative Science.

Rao, A. (2025). Enhancing credit scoring models with explainable AI and SHAP-based interpretability. Journal of Business Analytics and Financial Data Processing, 9(11), 1–11.

Roy, J. K., & Vasa, L. (2025). Transforming credit risk assessment: A systematic review of AI and machine learning applications. Journal of Infrastructure, Policy and Development.

Schmitt, M., & Weber, R. (2024). Explainable automated machine learning for credit risk prediction. arXiv.

Shreya, & Pathak, H. (2025). Explainable Artificial Intelligence Credit Risk Assessment Using Machine Learning. ArXiv.

Torrent, N. L., Visani, G., & Bagli, E. (2020). PSD2 Explainable AI Model for Credit Scoring. ArXiv.