مجلة إدارة الأعمال والدراسات الاقتصادية
Volume 12, Numéro 1, Pages 135-158
2026-06-02
Machine Learning Based Customer Churn Prediction: A Bibliometric Analysis Of Techniques, Key Methodologies, And Sectoral Trends (2011-2025)
Authors : Yahiaoui Nour El Houda .
Abstract
In a competitive business landscape, predictive models leveraging machine learning (ML) are indispensable for customer churn prediction. In order to represent the field’s growth and future frontiers, the study analyzes 260 Scopus-indexed publications between 2011 and 2025 using RStudio and Biblioshiny. The data analysis reveals a dramatic shift away from the pre-2010 trends in comparative analyses and the basic methods of data mining into the dominant themes. Major focus is on Ensemble Classifiers such as XGBoost, Deep Learning architectures such as LSTM, and Big Data scalability models for challenging industries such as telecommunications, banking and e-commerce. The key journals for the major studies of high impact, IEEE Access, occupy significant positions; authors such as Ravi, V. and Vu, V.H. contributed to emergent clusters in support of predictive modeling. The regional patterns show India’s productivity and influence, followed closely by China and impactful research from South Korea. This paper demonstrates the critical need for more XAI and qualitative analysis, including capturing ethical consequences, business actionability, and behavioral details.
Keywords
Customer Churn Prediction ; Machine Learning (ML) ; Bibliometric Analysis ; Deep Learning ; Explainable AI (XAI)
Les articles similaires
Smail Mohamed Yacine
.
Asri Ayoub
.
pages 79-97.
Kalli Radhia
.
pages 192-207.
Boumedda-tiab Fahima
.
pages 395-406.
Nadia Allili
.
pages 1-23.
Benmahamed Houda
.
Mihoub Sameh
.
Toubal Ibtissem
.
pages 1-20.
