Customer Churn Prediction Using Machine Learning: A Review of Techniques, Imbalance Handling, Explainability, and Personalized Retention

Author

Mrs. Gorisetty. Nirosha, Saripalli Jabili


Abstract

 Customer churn prediction applies machine learning (ML) to identify customers who are likely to stop using a product or service, so that a business can act before they leave. This paper reviews fifteen studies published between 2013 and 2026 across banking, telecommunications, e-commerce, and industrial settings. It covers classical classifiers (Logistic Regression, Decision Tree, Random Forest, SVM), boosting and ensemble methods (XGBoost, weighted ensembles), deep models (Bi-LSTM, CNN/RNN, Transformer), social-network features, hard and soft data fusion, class-imbalance handling (SMOTE and density-regularized oversampling), explainable AI, and profit-based evaluation. The synthesis shows that preprocessing, imbalance handling, and the choice of evaluation metric influence outcomes as much as the choice of algorithm, and that most studies stop at predicting who will churn rather than explaining why and recommending what to do. To address this gap, the paper proposes a workflow that combines Logistic Regression, Random Forest, and XGBoost with SMOTE applied to the training data only, SHAP-based factor analysis, and rule-based personalized retention recommendations. The framework is a design proposal and is not experimentally evaluated in this paper; the paper closes with open challenges and future directions.


Keywords

Customer Churn Prediction, Machine Learning, Customer Retention, Predictive Analytics, Classification Algorithms, Class Imbalance, Explainable AI, Personalized Retention Strategies.



Full Text:

Download Paper PDF


References


[1] S. Y. Al-Sultan and I. A. Al-Baltah, “An Improved Random Forest Algorithm (ERFA) Utilizing an Unbalanced and Balanced Dataset to Predict Customer Churn in the Banking Sector,” IEEE Access, vol. 12, 2024, doi: 10.1109/ACCESS.2024.3395542.

[2] B. Amiri and S. H. Hosseini, “Unveiling the Power of Social Influence: A Machine Learning Framework for Churn Prediction With Network Analysis,” IEEE Access, vol. 12, pp. 71271–71285, 2024, doi: 10.1109/ACCESS.2024.3402684.

[3] S. M. Shrestha and A. Shakya, “A Customer Churn Prediction Model using XGBoost for the Telecommunication Industry in Nepal,” Procedia Computer Science, vol. 215, pp. 652–661, 2022, doi: 10.1016/j.procs.2022.12.067.

[4] A. Manzoor, M. A. Qureshi, E. Kidney and L. Longo, “A Review on Machine Learning Methods for Customer Churn Prediction and Recommendations for Business Practitioners,” IEEE Access, vol. 12, pp. 70434–70463, 2024, doi: 10.1109/ACCESS.2024.3402092.

[5] I. Ullah, B. Raza, A. K. Malik, M. Imran, S. U. Islam and S. W. Kim, “A Churn Prediction Model Using Random Forest: Analysis of Machine Learning Techniques for Churn Prediction and Factor Identification in Telecom Sector,” IEEE Access, vol. 7, pp. 60134–60149, 2019, doi: 10.1109/ACCESS.2019.2914999.

[6] S. Wu, W.-C. Yau, T.-S. Ong and S.-C. Chong, “Integrated Churn Prediction and Customer Segmentation Framework for Telco Business,” IEEE Access, vol. 9, pp. 62118–62136, 2021, doi: 10.1109/ACCESS.2021.3073776.

[7] M. Alizadeh, D. S. Zadeh, B. Moshiri and A. Montazeri, “Development of a Customer Churn Model for Banking Industry Based on Hard and Soft Data Fusion,” IEEE Access, vol. 11, pp. 29759–29768, 2023, doi: 10.1109/ACCESS.2023.3257352.

[8] A. Inkesar, “An Explainable Machine Learning Approach With Density-Regularized Oversampling for Industrial Customer Churn Prediction,” IEEE Access, vol. 14, pp. 119355–119383, 2026, doi: 10.1109/ACCESS.2026.3718205.

[9] Y. Li, Y. Tao, M. Zhu, Z. Chen, Z. Wen and B. Zhu, “TBformer: Multi-Scale Transformer With Time-Behavior Attention for Multi-Modal Customer Churn Prediction,” IEEE Trans. Consum. Electron., vol. 71, no. 2, pp. 3192–3203, May 2025, doi: 10.1109/TCE.2025.3563905.

[10] M. Afzal, S. Rahman, D. Singh and A. Imran, “Cross-Sector Application of Machine Learning in Telecommunications: Enhancing Customer Retention Through Comparative Analysis of Ensemble Methods,” IEEE Access, vol. 12, pp. 115256–115267, 2024, doi: 10.1109/ACCESS.2024.3445281.

[11] Q. Wei, “Personalized and Contextualized Data Analysis for E-Commerce Customer Retention Improvement With Bi-LSTM Churn Prediction,” IEEE Trans. Consum. Electron., vol. 71, no. 2, pp. 4406–4414, May 2025, doi: 10.1109/TCE.2024.3376672.

[12] N.-T. Lee, H.-C. Lee, J. Hsin and S.-H. Fang, “Prediction of Customer Behavior Changing via a Hybrid Approach,” IEEE Open J. Comput. Soc., vol. 5, pp. 27–38, 2024, doi: 10.1109/OJCS.2023.3336904.

[13] T. Verbraken, W. Verbeke and B. Baesens, “A Novel Profit Maximizing Metric for Measuring Classification Performance of Customer Churn Prediction Models,” IEEE Trans. Knowl. Data Eng., vol. 25, no. 5, pp. 961–973, May 2013, doi: 10.1109/TKDE.2012.50.

[14] I. N. M. Adiputra and P. Wanchai, “Customer Churn Prediction Using Weight Average Ensemble Machine Learning Model,” in Proc. 20th Int. Joint Conf. Computer Science and Software Engineering (JCSSE), 2023, pp. 90–94, doi: 10.1109/JCSSE58229.2023.10202105.

[15] M. Usman, W. Ahmad and A. Fong, “Design and Implementation of a System for Comparative Analysis of Learning Architectures for Churn Prediction,” IEEE Commun. Mag., vol. 59, no. 9, pp. 86–90, Sep. 2021, doi: 10.1109/MCOM.110.2100145.

 

 

Share your valuable work from Social Media Buttons