Predicting Loan Approval Decisions: A Comparative Machine Learning Study

DOI

10.2991/978-94-6239-801-6_155How to use a DOI?

Keywords

Loan approval prediction; Credit scoring; Supervised learning; Decision tree; Feature importance

Abstract

The experiment used the public Kaggle Loan Approval Prediction Dataset, which contains 4,269 applications. The paper treated loan status as a two-class target and kept the preprocessing workflow deliberately simple: removing the identifier field, checking missing values, encoding categorical variables, splitting the data into training and test samples, and scaling features for algorithms that are sensitive to variable magnitude. Five models were compared, namely Logistic Regression, Decision Tree, Random Forest, K-Nearest Neighbors, and Support Vector Machine. Accuracy, precision, recall, and F1-score were used to evaluate the hold-out test results. The two tree-based models produced the clearest performance, and the Decision Tree obtained the highest score in this run, with an accuracy of 0.9813 and an F1-score of 0.9849. Random Forest was only slightly lower. Its feature importance output showed that CIBIL score dominated the model decision process, while loan term and loan amount contributed at a much smaller level. The findings suggest that machine learning can support loan screening on structured data, but the result should be understood as a classroom-style model comparison rather than a ready-to-use banking system.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Wanxuan Sun
PY  - 2026
DA  - 2026/10/23
TI  - Predicting Loan Approval Decisions: A Comparative Machine Learning Study
BT  - Proceedings of the 2026 2nd International Conference on Financial Innovation and Marketing Management (FIMM 2026)
PB  - Atlantis Press
SP  - 1474
EP  - 1480
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6239-801-6_155
DO  - 10.2991/978-94-6239-801-6_155
ID  - Sun2026
ER  -