Enhancing Commercial Bank Customer Churn Prediction with Bayesian-Optimized Stacking Algorithm

DOI

10.2991/978-2-38476-487-7_12How to use a DOI?

Keywords

Customer Churn Prediction; Stacking; Bayesian Optimization; Commercial Bank

Abstract

This study tackles customer churn in the financial sector by introducing a Bayesian-optimized Stacking ensemble model to enhance prediction accuracy and stability. The proposed model incorporates five base classifiers and uses a Decision Tree with a maximum depth of 4 as the meta-learner in a two-layer Stacking structure. Bayesian optimization is employed to fine-tune hyperparameters, addressing key challenges like class imbalance and model generalization. Experiments on the Kaggle customer churn dataset demonstrate the model’s effectiveness, achieving a test F1-score of 0.9791 and significantly outperforming traditional single models and other ensemble methods. The main contributions of this work are the development of a Bayesian-optimized Stacking framework for improved churn prediction, and its validation on real-world financial data, demonstrating enhanced scalability, generalization, and practical value for banking applications.

Copyright

© 2025 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  - Zihua Mai
PY  - 2025
DA  - 2025/11/10
TI  - Enhancing Commercial Bank Customer Churn Prediction with Bayesian-Optimized Stacking Algorithm
BT  - Proceedings of the 2025 International Conference on Digital Technology and Educational Psychology (DTEP 2025)
PB  - Atlantis Press
SP  - 145
EP  - 151
SN  - 2352-5398
UR  - https://doi.org/10.2991/978-2-38476-487-7_12
DO  - 10.2991/978-2-38476-487-7_12
ID  - Mai2025
ER  -