Transformation of Traditional Models to AI: SLR on the Application of Machine Learning in Mortality Prediction

Vita Nuraini, Nina Fitriyati

Abstract


The application of machine learning (ML) in actuarial science and life insurance has driven digital transformation in mortality risk prediction. This article conducts research using the Systematic Literature Review (SLR) methodology with the PRISMA approach to evaluate the performance comparison between ML methods and traditional actuarial models in predicting mortality risk. This study analyzed publication trends, geographic and institutional distribution, and methodologies used in the literature published between 2019 and 2025. The results from SLR show that ML methods, especially Random Forest and XGBoost, have superior predictive accuracy compared to traditional actuarial models such as Traditional Logistic Regression and Cox Proportional Hazards. However, despite the obvious accuracy advantage, issues of interpretability and long-term stability remain a major challenge in implementing ML in the actuarial industry. This study also identifies the need for a hybrid approach combining the strengths of both methodologies to improve prediction accuracy while maintaining high interpretability. This study suggests the need for further development in the application of ML by the regulation and compliance of the insurance industry. The findings provide insights for actuarial practitioners, regulators, and academics regarding the potential and challenges of using ML in mortality risk prediction.

Keywords


Machine learning, actuarial traditional model, hybrid model, mortality risk

Full Text:

PDF

References


[1] S. Levantesi, A. Nigri, and G. Piscopo, “Longevity risk management through Machine Learning: state of the art,” Insur. Mark. Co., vol. 11, no. 1, pp. 11–20, 2020, doi: 10.21511/ins.11(1).2020.02.

[2] R. Lee and L. R. Carter, “Modeling and forecasting U. S. mortality,” J. Am. Stat. Assoc., vol. 87, pp. 659–671, 1992, [Online]. Available: https://api.semanticscholar.org/CorpusID:29804325

[3] R. Nopour, L. Erfannia, N. Mehrabi, M. Mashoufi, A. Mahdavi, and M. Shanbehzadeh, “Comparison of Two Statistical Models for Predicting Mortality in COVID-19 Patients in Iran,” Shiraz E Med. J., vol. 23, no. 6, 2022, doi: 10.5812/semj.119172.

[4] B. Avanzi, G. Taylor, M. Wang, and B. Wong, “Machine Learning with High-Cardinality Categorical Features in Actuarial Applications,” ASTIN Bull., vol. 54, no. 2, pp. 213–238, 2024, doi: 10.1017/asb.2024.7.

[5] N. Bishop, “Application of Machine Learning Techniques in Insurance Underwriting,” J. Actuar. Res., vol. 2, no. 1, pp. 1–13, 2024, doi: 10.47941/jar.1756.

[6] B. Mahohoho, C. Chimedza, F. Matarise, and S. Munyira, “Artificial Intelligence-Based Automated Actuarial Pricing and Underwriting Model for the General Insurance Sector,” Open J. Stat., vol. 14, no. 03, pp. 294–340, 2024, doi: 10.4236/ojs.2024.143014.

[7] M. J. Page et al., “The PRISMA 2020 statement: An updated guideline for reporting systematic reviews,” BMJ, vol. 372, 2021, doi: 10.1136/bmj.n71.

[8] E. von Elm, D. G. Altman, M. Egger, S. J. Pocock, P. C. Gøtzsche, and J. P. Vandenbroucke, “Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.,” BMJ, vol. 335, no. 7624, pp. 806–808, Oct. 2007, doi: 10.1136/bmj.39335.541782.AD.

[9] Y. Boo and Y. Choi, “Comparing logistic regression models with alternative machine learning methods to predict the risk of drug intoxication mortality,” Int. J. Environ. Res. Public Health, vol. 17, no. 3, 2020, doi: 10.3390/ijerph17030897.

[10] F. H. Bitew, S. H. Nyarko, L. Potter, and C. S. Sparks, “Machine learning approach for predicting under-five mortality determinants in Ethiopia: evidence from the 2016 Ethiopian Demographic and Health Survey,” Genus, vol. 76, no. 1, 2020, doi: 10.1186/s41118-020-00106-2.

[11] M. Maier, H. Carlotto, S. Saperstein, F. Sanchez, S. Balogun, and S. Merritt, “Improving the accuracy and transparency of underwriting with artificial intelligence to transform the life-insurance industry,” AI Mag., vol. 41, no. 3, pp. 78–93, 2020, doi: 10.1609/AIMAG.V41I3.5320.

[12] W. Lee et al., “Machine learning enhances the performance of short and long-term mortality prediction model in non-ST-segment elevation myocardial infarction,” Sci. Rep., vol. 11, no. 1, pp. 1–14, 2021, doi: 10.1038/s41598-021-92362-1.

[13] Y. Li, M. Chen, H. Lv, P. Yin, L. Zhang, and P. Tang, “A novel machine-learning algorithm for predicting mortality risk after hip fracture surgery,” Injury, vol. 52, no. 6, pp. 1487–1493, 2021, doi: 10.1016/j.injury.2020.12.008.

[14] J. Lee, J. Cai, F. Li, and Z. A. Vesoulis, “Predicting mortality risk for preterm infants using random forest,” Sci. Rep., vol. 11, no. 1, pp. 1–9, 2021, doi: 10.1038/s41598-021-86748-4.

[15] J. M. Brahmbhatt et al., “The lung allocation score and other available models lack predictive accuracy for post-lung transplant survival,” J. Hear. Lung Transplant., vol. 41, no. 8, pp. 1063–1074, 2022, doi: https://doi.org/10.1016/j.healun.2022.05.008.

[16] V. Garcia-Montemayor et al., “Predicting mortality in hemodialysis patients using machine learning analysis,” Clin. Kidney J., vol. 14, no. 5, pp. 1388–1395, 2020, doi: 10.1093/ckj/sfaa126.

[17] S. Sinha et al., “Comparison of machine learning techniques in prediction of mortality following cardiac surgery: analysis of over 220 000 patients from a large national database,” Eur. J. Cardio-Thoracic Surg., vol. 63, no. 6, p. ezad183, 2023, doi: 10.1093/ejcts/ezad183.

[18] K. McDonnell, F. Murphy, B. Sheehan, L. Masello, and G. Castignani, “Deep learning in insurance: Accuracy and model interpretability using TabNet,” Expert Syst. Appl., vol. 217, p. 119543, 2023, doi: https://doi.org/10.1016/j.eswa.2023.119543.

[19] Y. Shen, X. Yang, H. Liu, and Z. Li, “Advancing mortality rate prediction in European population clusters: integrating deep learning and multiscale analysis,” Sci. Rep., vol. 14, no. 1, pp. 1–16, 2024, doi: 10.1038/s41598-024-56390-x.

[20] Y. Chen and A. Q. M. Khaliq, “Comparative Study of Mortality Rate Prediction Using Data-Driven Recurrent Neural Networks and the Lee–Carter Model,” Big Data Cogn. Comput., vol. 6, no. 4, 2022, doi: 10.3390/bdcc6040134.

[21] C. Clemente, G. R. Guerreiro, and J. M. Bravo, “Modelling Motor Insurance Claim Frequency and Severity Using Gradient Boosting †,” Risks, vol. 11, no. 9, pp. 1–20, 2023, doi: 10.3390/risks11090163.

[22] I. Vagliano et al., “Can we reliably automate clinical prognostic modelling? A retrospective cohort study for ICU triage prediction of in-hospital mortality of COVID-19 patients in the Netherlands.,” Int. J. Med. Inform., vol. 160, p. 104688, Apr. 2022, doi: 10.1016/j.ijmedinf.2022.104688.

[23] K. Nakamura et al., “Risk of Mortality Prediction Involving Time-Varying Covariates for Patients with Heart Failure Using Deep Learning,” Diagnostics, vol. 12, no. 12, pp. 1–14, 2022, doi: 10.3390/diagnostics12122947.

[24] B. Nistal-Nuño, “Machine learning applied to a Cardiac Surgery Recovery Unit and to a Coronary Care Unit for mortality prediction,” J. Clin. Monit. Comput., vol. 36, no. 3, pp. 751–763, 2022, doi: 10.1007/s10877-021-00703-2.

[25] A. H. T. Chia et al., “Explainable machine learning prediction of ICU mortality,” Informatics Med. Unlocked, vol. 25, p. 100674, 2021, doi: https://doi.org/10.1016/j.imu.2021.100674.

[26] F. S. Chou and L. V. Ghimire, “Machine Learning for Mortality Prediction in Pediatric Myocarditis,” Front. Pediatr., vol. 9, no. April, pp. 1–8, 2021, doi: 10.3389/fped.2021.644922.

[27] R. R. Lopes et al., “Temporal validation of 30-day mortality prediction models for transcatheter aortic valve implantation using statistical process control – An observational study in a national population,” Heliyon, vol. 9, no. 6, p. e17139, 2023, doi: https://doi.org/10.1016/j.heliyon.2023.e17139.

[28] J. L. Andrade and J. L. Valencia, “A Fuzzy Random Survival Forest for Predicting Lapses in Insurance Portfolios Containing Imprecise Data,” Mathematics, vol. 11, no. 1, 2023, doi: 10.3390/math11010198.

[29] D. Kovacs, D. R. Msanga, S. E. Mshana, M. Bilal, K. Oravcova, and L. Matthews, “Developing practical clinical tools for predicting neonatal mortality at a neonatal intensive care unit in Tanzania,” BMC Pediatr., vol. 21, no. 1, pp. 1–10, 2021, doi: 10.1186/s12887-021-03012-4.

[30] J. C. Penny-Dimri, C. Bergmeir, C. M. Reid, J. Williams-Spence, L. A. Perry, and J. A. Smith, “Tree-based survival analysis improves mortality prediction in cardiac surgery,” Front. Cardiovasc. Med., vol. 10, no. July, pp. 1–8, 2023, doi: 10.3389/fcvm.2023.1211600.

[31] M. Huber, P. Schober, S. Petersen, and M. M. Luedi, “Decision curve analysis confirms higher clinical utility of multi-domain versus single-domain prediction models in patients with open abdomen treatment for peritonitis,” BMC Med. Inform. Decis. Mak., vol. 23, no. 1, pp. 1–12, 2023, doi: 10.1186/s12911-023-02156-w.

[32] A. A. Wilson, A. Nehme, A. Dhyani, and K. Mahbub, “A Comparison of Generalised Linear Modelling with Machine Learning Approaches for Predicting Loss Cost in Motor Insurance,” Risks, vol. 12, no. 4, 2024, doi: 10.3390/risks12040062.

[33] S. M. Kagerbauer et al., “Susceptibility of AutoML mortality prediction algorithms to model drift caused by the COVID pandemic,” BMC Med. Inform. Decis. Mak., vol. 24, no. 1, pp. 1–13, 2024, doi: 10.1186/s12911-024-02428-z.




DOI: https://doi.org/10.18860/cauchy.v10i2.35972

Refbacks

  • There are currently no refbacks.


Copyright (c) 2025 Vita Nuraini, Nina Fitriyati

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Editorial Office
Mathematics Department,
Universitas Islam Negeri Maulana Malik Ibrahim Malang
Gajayana Street 50 Malang, East Java, Indonesia 65144
Faximile (+62) 341 558933
e-mail: cauchy@uin-malang.ac.id

Creative Commons License
CAUCHY: Jurnal Matematika Murni dan Aplikasi is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.