Implementation of Machine Learning Algorithms for Customer Churn Prediction to Enhance Customer Retention

DOI

10.2991/978-94-6239-604-3_52How to use a DOI?

Keywords

Customer churn; banking; machine learning; LightGBM; feature importance; VINTAGE; retention strategy; predictive model

Abstract

Customer churn is a major issue in the highly competitive banking sector, directly affecting long-term profitability and business sustainability. This study presents a comprehensive framework for developing and evaluating a predictive customer churn model aimed at supporting the implementation of more proactive and effective customer retention strategies. The methodology encompasses several systematic stages, including in-depth data exploration, meticulous data preprocessing, strategic feature engineering, and the evaluation of various machine learning algorithms.

Among the models tested, including logistic regression, decision tree, random forest, and gradient boosting, LightGBM demonstrated the best performance across multiple evaluation metrics. Its superior capability in handling complex data structures and capturing non-linear patterns underscores its effectiveness in distinguishing between customers likely to churn and those who are not. Explainability analysis using SHAP identified BALANCE, NOA_DORMANT, and CA_WAIR as the most influential predictors, with higher dormant account activity increasing churn likelihood, while stronger financial engagement indicators generally reduced risk. Multicollinearity issues were addressed using Variance Inflation Factor (VIF) analysis, resulting in the removal of several highly collinear regional variables while retaining key financial features with stable VIF values.

The resulting predictive model provides actionable insights into customer behavior and offers significant potential for enabling more targeted, personalized, and cost-effective retention strategies, supporting improved customer loyalty and sustainable business growth within the banking industry.

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  - Jesvika Sari Melyana Sihombing
AU  - Muhammad Zarlis
PY  - 2026
DA  - 2026/02/26
TI  - Implementation of Machine Learning Algorithms for Customer Churn Prediction to Enhance Customer Retention
BT  - Proceedings of the 5th International Conference on Economic Development and Business Culture (ICEDBC 2025)
PB  - Atlantis Press
SP  - 497
EP  - 517
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6239-604-3_52
DO  - 10.2991/978-94-6239-604-3_52
ID  - Sihombing2026
ER  -