Binary Firefly Algorithm-Based Optimization of TabNet and DNN Models for Customer Churn Prediction
DOI:
https://doi.org/10.62411/faith.3048-3719-376Keywords:
Binary Firefly Algorithm, Customer Churn Prediction, Deep Learning; Feature Selection, Predictive Analytics, TabNet, Telecommunications, Tabular Data LearningAbstract
Customer churn remains a major challenge in the telecommunications industry, where intense market competition and evolving customer expectations significantly affect customer retention. Accurate churn prediction is therefore essential for developing effective retention strategies. However, conventional machine learning approaches often exhibit limited capability in capturing complex nonlinear relationships and may be sensitive to high-dimensional, noisy, and redundant features commonly found in telecom datasets. To address these challenges, this study investigates the effectiveness of Binary Firefly Algorithm (BFA)-based feature optimization for deep learning-driven customer churn prediction. BFA is employed to identify the most informative feature subset, reducing feature redundancy and improving model efficiency. Three deep learning architectures are evaluated using the Maven Analytics Telecom Customer Churn dataset, namely a Deep Neural Network (DNN), TabNet, and a hybrid TabNet–DNN model. The DNN is designed to learn complex nonlinear patterns, whereas TabNet utilizes a sequential attention mechanism to perform adaptive feature representation learning for tabular data. Experimental results demonstrate that BFA-based feature optimization substantially improves predictive performance across all evaluated models. Among the investigated architectures, TabNet achieved the best overall performance, obtaining an accuracy of 96.42% and an AUROC of 0.9885. Although the hybrid TabNet–DNN model achieved competitive results, it did not consistently outperform the standalone TabNet model. These findings suggest that effective feature optimization contributes more significantly to predictive performance than increased architectural complexity in structured telecom churn prediction tasks. The study further highlights the potential of BFA-enhanced deep learning models as a robust framework for customer churn analytics in telecommunications.
Downloads
References
S. M. Kostić, M. I. Simić, and M. V. Kostić, “Social Network Analysis and Churn Prediction in Telecommunications Using Graph Theory,” Entropy, vol. 22, no. 7, p. 753, Jul. 2020, doi: 10.3390/e22070753.
M. J. Nur, D. R. I. Moses Setiadi, A. A. Ojugo, and M. T. Nguyen, “Improving Customer Churn Prediction Using Domain-Driven Feature Engineering, Resampling, and CatBoost with Explainability Extensions,” in 2025 International Seminar on Application for Technology of Information and Communication (iSemantic), Sep. 2025, pp. 493–499. doi: 10.1109/ISemantic67418.2025.11291801.
N. R. Al-Molhem, Y. Rahal, and M. Dakkak, “Social network analysis in Telecom data,” J. Big Data, vol. 6, no. 1, p. 99, Dec. 2019, doi: 10.1186/s40537-019-0264-6.
M. Loukili, “Supervised Learning Algorithms for Predicting Customer Churn with Hyperparameter Optimization,” Int. J. Adv. Soft Comput. its Appl., vol. 14, no. 3, pp. 50–63, Dec. 2022, doi: 10.15849/IJASCA.221128.04.
A. Prashanthan, “An Integrated Framework for Optimizing Customer Retention Budget using Clustering, Classification, and Mathematical Optimization,” J. Comput. Theor. Appl., vol. 3, no. 1, pp. 45–63, Jul. 2025, doi: 10.62411/jcta.13194.
T. Xu, Y. Ma, and K. Kim, “Telecom Churn Prediction System Based on Ensemble Learning Using Feature Grouping,” Appl. Sci., vol. 11, no. 11, p. 4742, May 2021, doi: 10.3390/app11114742.
A. Bugajev, R. Kriauzienė, O. Vasilecas, and V. Chadyšas, “The Impact of Churn Labelling Rules on Churn Prediction in Telecommunications,” Informatica, pp. 247–277, May 2022, doi: 10.15388/22-INFOR484.
D. K. Sahoo and P. K. Sahoo, “Efficiency, productivity dynamics and determinants of productivity growth in Indian telecommunication industries: An empirical analysis,” J. Public Aff., vol. 22, no. 1, Feb. 2022, doi: 10.1002/pa.2353.
E. Xevelonakis and P. Som, “The impact of social network-based segmentation on customer loyalty in the telecommunication industry,” J. Database Mark. Cust. Strateg. Manag., vol. 19, no. 2, pp. 98–106, Jun. 2012, doi: 10.1057/dbm.2012.12.
R. E. Ako et al., “Effects of Data Resampling on Predicting Customer Churn via a Comparative Tree-based Random Forest and XGBoost,” J. Comput. Theor. Appl., vol. 2, no. 1, pp. 86–101, Jun. 2024, doi: 10.62411/jcta.10562.
Y. Khan, S. Shafiq, A. Naeem, S. Ahmed, N. Safwan, and S. Hussain, “Customers Churn Prediction using Artificial Neural Networks (ANN) in Telecom Industry,” Int. J. Adv. Comput. Sci. Appl., vol. 10, no. 9, 2019, doi: 10.14569/IJACSA.2019.0100918.
J. Nagaraju and J. Vijaya, “Boost customer churn prediction in the insurance industry using meta-heuristic models,” Int. J. Inf. Technol., vol. 14, no. 5, pp. 2619–2631, Aug. 2022, doi: 10.1007/s41870-022-01017-5.
V. P. Malikireddy and M. Kasa, “Customer Churns Prediction Model Based on Machine Learning Techniques: A Systematic Review,” in Proceedings of the 3rd International Conference on Integrated Intelligent Computing Communication & Security (ICIIC 2021), 2021. doi: 10.2991/ahis.k.210913.021.
S. W. Fujo, S. Subramanian, and M. A. Khder, “Customer Churn Prediction in Telecommunication Industry Using Deep Learning,” Inf. Sci. Lett., vol. 11, no. 1, pp. 185–198, Jan. 2022, doi: 10.18576/isl/110120.
S. Ö. Arik and T. Pfister, “TabNet: Attentive Interpretable Tabular Learning,” Proc. AAAI Conf. Artif. Intell., vol. 35, no. 8, pp. 6679–6687, May 2021, doi: 10.1609/aaai.v35i8.16826.
S. I. Ele, U. R. Alo, H. F. Nweke, and O. A. Ofem, “Regression-Based Machine Learning Framework for Customer Churn Prediction in Telecommunication Industry,” J. Adv. Inf. Technol., vol. 14, no. 5, pp. 1046–1055, 2023, doi: 10.12720/jait.14.5.1046-1055.
V. Chang, K. Hall, Q. Xu, F. Amao, M. Ganatra, and V. Benson, “Prediction of Customer Churn Behavior in the Telecommunication Industry Using Machine Learning Models,” Algorithms, vol. 17, no. 6, p. 231, May 2024, doi: 10.3390/a17060231.
P. Lalwani, M. K. Mishra, J. S. Chadha, and P. Sethi, “Customer churn prediction system: a machine learning approach,” Computing, vol. 104, no. 2, pp. 271–294, Feb. 2022, doi: 10.1007/s00607-021-00908-y.
D. Asif, M. S. Arif, and A. Mukheimer, “A data-driven approach with explainable artificial intelligence for customer churn prediction in the telecommunications industry,” Results Eng., vol. 26, p. 104629, Jun. 2025, doi: 10.1016/j.rineng.2025.104629.
S. S. Poudel, S. Pokharel, and M. Timilsina, “Explaining customer churn prediction in telecom industry using tabular machine learning models,” Mach. Learn. with Appl., vol. 17, p. 100567, Sep. 2024, doi: 10.1016/j.mlwa.2024.100567.
A. Prashanthan, R. Roshan, and M. Maduranga, “RetenNet: A Deployable Machine Learning Pipeline with Explainable AI and Prescriptive Optimization for Customer Churn Management,” J. Futur. Artif. Intell. Technol., vol. 2, no. 2, pp. 182–201, Jun. 2025, doi: 10.62411/faith.3048-3719-110.
P. Ndabarishye and A. K. Singh, “Understanding Customer Churn in Retail Banking through Explainable Predictive Analytics: Evidence of a Product Paradox,” J. Comput. Theor. Appl., vol. 3, no. 4, pp. 473–486, Apr. 2026, doi: 10.62411/jcta.15870.
S. K. Wagh, A. A. Andhale, K. S. Wagh, J. R. Pansare, S. P. Ambadekar, and S. H. Gawande, “Customer churn prediction in telecom sector using machine learning techniques,” Results Control Optim., vol. 14, p. 100342, Mar. 2024, doi: 10.1016/j.rico.2023.100342.
S. Saha, C. Saha, M. M. Haque, M. G. R. Alam, and A. Talukder, “ChurnNet: Deep Learning Enhanced Customer Churn Prediction in Telecommunication Industry,” IEEE Access, vol. 12, pp. 4471–4484, 2024, doi: 10.1109/ACCESS.2024.3349950.
Deepika Kumari, Santosh Kumar Singh, Sanjay Subhash Katira, Inumarthi V Srinivas, and Uday Salunkhe, “Telecom Customer Churn Forecasting Using Machine Learning: A Data-Driven Predictive Framework,” Metall. Mater. Eng., vol. 31, no. 4, pp. 922–929, Apr. 2025, doi: 10.63278/1536.
K. Mourad and R. Boudour, “A Modified Binary Firefly Algorithm to Solve Hardware/Software Partitioning Problem,” Informatica, vol. 45, no. 7, Nov. 2021, doi: 10.31449/inf.v45i7.3408.
A. G. Toprak, Ö. Berfin Mercan, and M. S. Osmanca, “ML-Based Telecom Customer Churn Analysis,” in 2025 33rd Signal Processing and Communications Applications Conference (SIU), Jun. 2025, pp. 1–4. doi: 10.1109/SIU66497.2025.11111814.
S. Saleh and S. Saha, “Customer retention and churn prediction in the telecommunication industry: a case study on a Danish university,” SN Appl. Sci., vol. 5, no. 7, p. 173, Jul. 2023, doi: 10.1007/s42452-023-05389-6.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Philip O. Odion, Maaruf M. Lawal, Sonia S. Joshua, Abdulrashid Abdulrauf
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
