{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T06:03:53Z","timestamp":1789625033468,"version":"build-2803163510"},"reference-count":26,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T00:00:00Z","timestamp":1789603200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Customer churn prediction in retail banking poses significant financial challenges, demanding models that are both accurate and interpretable under regulatory constraints. Existing approaches often suffer from data leakage, poor probability calibration, and inadequate handling of class imbalance.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      We present a calibrated, cost-sensitive, and explainable ensemble learning pipeline evaluated on the publicly available Bank Customer Churn dataset (ChurnModelling.csv,\n                      <jats:italic>n<\/jats:italic>\n                      = 10, 000). Seven ensemble classifiers - Random Forest, Extra-Trees, XGBoost, LightGBM, CatBoost, Stacking, and Voting - alongside a Logistic Regression baseline were compared using a strictly leakage-safe design incorporating class-weighting, decision-threshold optimization, isotonic probability calibration, Bayesian hyperparameter optimization via Optuna, and repeated stratified cross-validation with Friedman and Nemenyi post hoc testing.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The Bayesian-tuned LightGBM achieved the highest CV ROC-AUC of 0.8649 \u00b1 0.0096 and test ROC-AUC of 0.8668, PR-AUC of 0.7139, and Brier score of 0.1395. At the cost-sensitive threshold (0.265), recall reached 0.7592 with precision of 0.5091. Cumulative gains analysis showed that targeting the top 40% of customers by predicted churn probability captures approximately 80% of actual churners.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>The pipeline demonstrates that class-weighting combined with threshold optimization is a calibration-preserving alternative to SMOTE oversampling. TreeSHAP analysis identifies Age, NumOfProducts, and IsActiveMember as the most influential predictors. Limitations include reliance on a single static dataset and the exploratory nature of the statistical significance claims. The framework is presented as a reproducible, leakage-safe single-dataset case study rather than a validated general solution.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/frai.2026.1889639","type":"journal-article","created":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T05:33:39Z","timestamp":1789623219000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Calibrated, cost-sensitive, explainable ensemble learning for customer churn prediction in retail banking: a leakage-safe pipeline and single-dataset case study"],"prefix":"10.3389","volume":"9","author":[{"given":"Farhana","family":"Karim","sequence":"first","affiliation":[{"name":"Harrison College of Business and Computing, Southeast Missouri State University","place":["Cape Girardeau, MO, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Md. Ishtiaque","family":"Alam","sequence":"additional","affiliation":[{"name":"Orfalea College of Business, California Polytechnic State University","place":["San Luis Obispo, CA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aram","family":"Lalafaryan","sequence":"additional","affiliation":[{"name":"Department of Business Administration, Westcliff University","place":["Irvine, CA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Artyom","family":"Mkrtchyan","sequence":"additional","affiliation":[{"name":"Department of Business Administration, Stanton University","place":["Los Angeles, CA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anzhela","family":"Hovakimyan","sequence":"additional","affiliation":[{"name":"Department of Business Administration, Merit University","place":["Los Angeles, CA, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Md","family":"Mustafizur","sequence":"additional","affiliation":[{"name":"Department of Business Administration, University of the Potomac","place":["Washington, DC, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gazi Touhidul","family":"Alam","sequence":"additional","affiliation":[{"name":"College of Graduate and Professional Studies, Trine University","place":["Detroit, MI, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mia Md. Tofayel Gonee","family":"Manik","sequence":"additional","affiliation":[{"name":"Department of Information Technology, University of the Cumberlands","place":["Williamsburg, KY, United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,17]]},"reference":[{"key":"B1","first-page":"135","article-title":"\u201cRelationship between angiotensin converting enzyme gene and cardiac autonomic neuropathy among australian population,\u201d","volume-title":"International Conference on Health Information Science (HIS), Lecture Notes in Computer Science","author":"Abdalrada","year":"2018"},{"key":"B2","first-page":"2623","article-title":"\u201cOptuna: a next-generation hyperparameter optimization framework,\u201d","volume-title":"Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Akiba","year":"2019"},{"key":"B3","first-page":"12","article-title":"\u201cAnalysis of breast cancer survivability using machine learning predictive technique for post-surgical patients,\u201d","volume-title":"Proceedings of the International Conference for ICT (ICICT)-Zambia","author":"Al-Quraishi","year":"2023"},{"key":"B4","first-page":"6679","article-title":"\u201cTabNet: Attentive interpretable tabular learning,\u201d","volume-title":"Proceedings of the AAAI Conference on Artificial Intelligence","author":"Arik","year":"2021"},{"key":"B5","doi-asserted-by":"publisher","first-page":"106","DOI":"10.1186\/1471-2105-14-106","article-title":"SMOTE for high-dimensional class-imbalanced data","volume":"14","author":"Blagus","year":"2013","journal-title":"BMC Bioinform"},{"key":"B6","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn"},{"key":"B7","doi-asserted-by":"publisher","first-page":"321","DOI":"10.1613\/jair.953","article-title":"SMOTE: Synthetic minority over-sampling technique","volume":"16","author":"Chawla","year":"2002","journal-title":"J. Artif. Intellig. Res"},{"key":"B8","doi-asserted-by":"crossref","first-page":"785","DOI":"10.1145\/2939672.2939785","article-title":"\u201cXGBoost: a scalable tree boosting system,\u201d","volume-title":"Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Chen","year":"2016"},{"key":"B9","first-page":"1","article-title":"Statistical comparisons of classifiers over multiple data sets","volume":"7","author":"Dem\u0161ar","year":"2006","journal-title":"J. Mach. Learn. Res"},{"key":"B10","unstructured":"Dhakad\n              S.\n            \n          \n          ChurnModelling.csv\n          San Francisco\n          Kaggle\n          Bank Customer Churn Prediction Dataset\n          \n          2020"},{"key":"B11","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-319-98074-4","volume-title":"Learning from Imbalanced Data Sets","author":"Fern\u00e1ndez","year":"2018"},{"key":"B12","doi-asserted-by":"publisher","first-page":"50","DOI":"10.1609\/aimag.v38i3.2741","article-title":"European union regulations on algorithmic decision-making and a \u201cright to explanation\u201d","volume":"38","author":"Goodman","year":"2017","journal-title":"AI Magazine"},{"key":"B13","first-page":"18932","article-title":"\u201cRevisiting deep learning models for tabular data,\u201d","volume-title":"Advances in Neural Information Processing Systems","author":"Gorishniy","year":"2021"},{"key":"B14","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/2382577.2382579","article-title":"Leakage in data mining: Formulation, detection, and avoidance","volume":"6","author":"Kaufman","year":"2012","journal-title":"ACM Trans. Knowl. Discov. Data"},{"key":"B15","first-page":"30","article-title":"\u201cLightGBM: a highly efficient gradient boosting decision tree,\u201d","volume-title":"Advances in Neural Information Processing Systems","author":"Ke","year":"2017"},{"key":"B16","author":"Kotler","year":"2016","journal-title":"Marketing Management"},{"key":"B17","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1038\/s42256-019-0138-9","article-title":"From local explanations to global understanding with explainable AI for trees","volume":"2","author":"Lundberg","year":"2020","journal-title":"Nat. Mach. Intellig"},{"key":"B18","article-title":"\u201cA unified approach to interpreting model predictions,\u201d","volume-title":"Advances in Neural Information Processing Systems","author":"Lundberg","year":"2017"},{"key":"B19","doi-asserted-by":"publisher","first-page":"1063","DOI":"10.1016\/j.procs.2024.05.140","article-title":"Online concept drift detector: Optimally balancing delay detection, runtime, memory, and accuracy","volume":"237","author":"Mahdi","year":"2024","journal-title":"Procedia Comp. Sci"},{"key":"B20","first-page":"1273","article-title":"\u201cCommunication-efficient learning of deep networks from decentralized data,\u201d","volume-title":"Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS)","author":"McMahan","year":"2017"},{"key":"B21","doi-asserted-by":"crossref","first-page":"625","DOI":"10.1145\/1102351.1102430","article-title":"\u201cPredicting good probabilities with supervised learning,\u201d","volume-title":"Proceedings of the 22nd International Conference on Machine Learning (ICML)","author":"Niculescu-Mizil","year":"2005"},{"key":"B22","first-page":"61","article-title":"\u201cProbabilistic outputs for support vector machines and comparisons to regularized likelihood methods,\u201d","volume-title":"Advances in Large Margin Classifiers","author":"Platt","year":"1999"},{"key":"B23","article-title":"\u201cCatBoost: Unbiased boosting with categorical features,\u201d","volume-title":"Advances in Neural Information Processing Systems","author":"Prokhorenkova","year":"2018"},{"key":"B24","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1016\/j.ejor.2011.09.031","article-title":"New insights into churn prediction in the telecommunication sector: a profit driven data mining approach","volume":"218","author":"Verbeke","year":"2012","journal-title":"Eur. J. Operat. Res"},{"key":"B25","doi-asserted-by":"publisher","first-page":"241","DOI":"10.1016\/S0893-6080(05)80023-1","article-title":"Stacked generalization","volume":"5","author":"Wolpert","year":"1992","journal-title":"Neural Netw"},{"key":"B26","first-page":"694","article-title":"\u201caccurate multiclass probability estimates,\u201d","volume-title":"Proceedings of the 8th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining","author":"Zadrozny","year":"2002"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1889639\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,17]],"date-time":"2026-09-17T05:33:45Z","timestamp":1789623225000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1889639\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,17]]},"references-count":26,"alternative-id":["10.3389\/frai.2026.1889639"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1889639","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,17]]},"article-number":"1889639"}}