Analysis of Customer Churn Classification for Sinarmas Syariah Lhokseumawe Insurance Services Using Deep Learning Tabnet and Explainable AI
DOI:
https://doi.org/10.30871/jaic.v10i3.12864Keywords:
Deep Learning, Tabnet, Churn, Classification, XAI, SHAPAbstract
Customer churn is a major challenge in the insurance industry because it directly affects customer retention, business sustainability, and company profitability. Early identification of customers at risk of churn is therefore essential for developing effective retention strategies. This study proposes an interpretable customer churn prediction framework for Sinarmas Syariah Lhokseumawe Insurance Services by integrating TabNet deep learning with Shapley Additive Explanations (SHAP). The dataset consists of 2,000 customer records containing demographic information, insurance transactions, premium payments, claims history, and customer interaction data. Due to the imbalanced class distribution, the Synthetic Minority Oversampling Technique (SMOTE) was applied exclusively to the training dataset to improve model learning while preventing data leakage. Model performance was evaluated using Accuracy, Precision, Recall, F1-Score, and ROC-AUC metrics. The experimental results demonstrate that the proposed approach achieved an accuracy of 98.77%, precision of 86.67%, recall of 92.86%, F1-score of 89.66%, and ROC-AUC of 0.995, indicating excellent classification performance. Furthermore, SHAP analysis revealed that communication, premi_2025, and reason_to_purchase were the most influential features affecting churn predictions. These findings highlight the importance of customer engagement, premium management, and purchasing motivations in customer retention. The proposed TabNet-SHAP framework provides both high predictive performance and model interpretability, making it a valuable decision-support tool for customer retention strategies in the insurance sector.
Downloads
References
[1] X. Liu, G. Xia, and X. Zhang, “Customer Churn Prediction Model Based on Hybrid Neural Networks,” Sci. Rep., vol. 14, p. 30707, 2024.
[2] T. P. Quinn and others, “A Primer on the Use of Machine Learning to Distil Knowledge from Data in Biological Psychiatry,” Mol. Psychiatry, vol. 29, no. 2, 2024.
[3] P. C. Verhoef and others, “Digital Transformation and Customer Engagement in Service Industries,” J. Bus. Res., vol. 122, pp. 889–901, 2021.
[4] R. T. Rust, “The Future of Service Management and Customer Retention,” J. Serv. Res., vol. 24, no. 1, pp. 3–10, 2021.
[5] S. S. Alam and others, “Customer Loyalty and Retention in Islamic Insurance Services,” J. Islam. Mark., vol. 13, no. 4, pp. 1010–1028, 2022.
[6] M. Bahri and others, “Machine Learning Applications for Insurance Customer Retention,” Expert Syst. Appl., vol. 223, p. 119845, 2023.
[7] Y. Huang and J. Chen, “Customer Churn Prediction in Financial Services Using Machine Learning Techniques,” Appl. Soft Comput., vol. 122, p. 108839, 2022.
[8] M. Ahmad and others, “Deep Learning Approaches for Customer Churn Prediction: A Systematic Review,” IEEE Access, vol. 11, pp. 23451–23470, 2023.
[9] H. Jabbar and R. Khan, “Comparative Analysis of Machine Learning Models for Churn Prediction,” Inf. Sci. (Ny)., vol. 608, pp. 1110–1128, 2022.
[10] C. Molnar, “Interpretable Machine Learning in Finance and Business Analytics,” Mach. Learn. Knowl. Extr., vol. 4, no. 2, pp. 245–268, 2022.
[11] A. Khaidar, Nurdin, Fajriana, Taufiq, and D. Hamdhana, “Classification Analysis of Single Tuition Fees Using the Random Forest Method with K-Fold Cross Validation,” J. Appl. Informatics Comput., vol. 10, no. 1, pp. 125–133, Feb. 2026, doi: 10.30871/jaic.v10i1.11798.
[12] S. M. Lundberg and others, “From Local Explanations to Global Understanding with Explainable AI for Trees,” Nat. Mach. Intell., vol. 2, no. 1, pp. 56–67, 2020.
[13] L. Merrick and A. Taly, “The Role of Explainable AI in Financial Decision Support Systems,” AI Mag., vol. 43, no. 2, pp. 125–138, 2022.
[14] S. O. Arik and T. Pfister, “TabNet: Attentive Interpretable Tabular Learning,” in Proceedings of the AAAI Conference on Artificial Intelligence, 2021, pp. 6679–6687.
[15] S. Wa, X. Lu, and M. Wang, “Stable and Interpretable Deep Learning for Tabular Data: Introducing InterpreTabNet,” arXiv Prepr. arXiv2310.02870, 2023.
[16] J. Si, W. Y. Cheng, and M. Cooper, “InterpreTabNet: Distilling Predictive Signals from Tabular Data by Salient Feature Interpretation,” arXiv Prepr. arXiv2406.00426, 2024.
[17] A. F. Tjahjono, H. Hasan, R. P. Putera, D. M. P. Indranto, and A. T. Hermawan, “Klasifikasi URL Phishing untuk SIEM: Perbandingan Model Machine Learning XGBoost dan Deep Learning TabNet dalam Deteksi Ancaman Siber,” Sains Data J. Stud. Mat. dan Teknol., vol. 3, no. 2, pp. 62–71, 2025, doi: 10.52620/sainsdata.v3i2.227.
[18] K. Perawatan, P. Pada, D. Tabular, E. Ismanto, A. Fadlil, and A. Yudhana, “Jurnal Computer Science and Information Technology ( CoSciTech ) Analisis Perbandingan Model Fully Connected Neural Networks ( FCNN ) dan TabNet Untuk,” J. Comput. Sci. Inf. Technol., vol. 5, no. 3, pp. 526–532, 2024.
[19] S. E. Awan, M. Bennamoun, F. Sohel, F. Sanfilippo, and G. Dwivedi, “A Reinforcement Learning-Based Approach for Imputing Missing Data,” Neural Comput. Appl., vol. 34, pp. 9701–9716, 2022, doi: 10.1007/s00521-022-06958-3.
[20] F. Pargent, F. Pfisterer, J. Thomas, and B. Bischl, “Regularized Target Encoding Outperforms Traditional Methods in Supervised Machine Learning with High Cardinality Features,” Comput. Stat., vol. 37, pp. 2671–2692, 2022, doi: 10.1007/s00180-022-01207-6.
[21] H. Kim, J. Lee, and others, “Data Preprocessing Combination to Improve the Performance of Quality Classification in the Manufacturing Process,” Electronics, vol. 11, no. 3, p. 477, 2022, doi: 10.3390/electronics11030477.
[22] M. Safii, Husain, Ika Okta Kirana, Sasha Aiko Leana, and Yuli Indahwati Gultom, “Model Prediksi Penjadwalan Produksi Energi Terbarukan dengan Algoritma XGBoost dan Analisis Interpretatif Menggunakan SHAP,” J. Sist. Inf. Triguna Dharma (JURSI TGD), vol. 4, no. 4, pp. 794–801, 2025, doi: 10.53513/jursi.v4i4.11443.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Syarifah Muliana, Taufiq Taufiq , Munirul Ula
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License (Attribution-ShareAlike 4.0 International (CC BY-SA 4.0) ) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
