Comparative Evaluation of Machine Learning Models with Explainability and Fairness for Diabetes Risk Prediction
- Vanya Dwi Nabila Universitas Sriwijaya, Palembang, Indonesia
-
Bayu Wijaya Putra
*
Universitas Sriwijaya, Palembang,
Indonesia
- M Rudi Sanjaya Universitas Sriwijaya, Palembang, Indonesia
- Dwi Rosa Indah Universitas Sriwijaya, Palembang, Indonesia (*) Corresponding Author
Abstract
Diabetes mellitus remains a major global health challenge, making early risk prediction essential for timely intervention and prevention. This study proposes a trustworthy machine learning framework for diabetes risk prediction by integrating predictive performance, explainability, and fairness evaluation. Six classification algorithms Logistic Regression, Decision Tree, Random Forest, XGBoost, LightGBM, and CatBoost were comparatively evaluated, with LightGBM selected as the best baseline model. Hyperparameter optimization was subsequently performed using Optuna with macro F1-score as the optimization objective to better address the imbalanced multiclass nature of the dataset. Model performance was assessed using Accuracy, Balanced Accuracy, Precision, Recall, F1-score, and Receiver Operating Characteristic–Area Under the Curve (ROC-AUC). Model explainability was analyzed using SHapley Additive exPlanations (SHAP), while fairness was evaluated using Fairlearn based on gender and race through Demographic Parity Difference and Equalized Odds Difference.Experimental results show that hyperparameter optimization increased Balanced Accuracy from 0.3755 to 0.4891 and macro F1-score from 0.3797 to 0.4333, indicating improved recognition of minority classes. Although overall Accuracy decreased from 0.8357 to 0.7023, this trade-off reflects a more balanced classification across diabetes categories, which is preferable for imbalanced clinical datasets where identifying minority cases is essential for early risk detection. SHAP analysis identified Body Mass Index (BMI), Age, Physical Health Days, Mental Health Days, and Income Level as the most influential predictors. Fairness evaluation further demonstrated low demographic disparities across gender and race. These findings demonstrate that integrating predictive performance, explainability, and fairness enables the development of a more transparent, equitable, and clinically reliable diabetes risk prediction framework.
Downloads
References
D. Magliano and E. J. Boyko, IDF Diabetes Atlas, 11th ed. Brussels, Belgium: International Diabetes Federation, 2025.
T. Hastie, R. Tibshirani, and J. H. Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd ed. New York, NY, USA: Springer, 2017.
D. Das, Aayushman, S. Kumar, M. A. Hussain, and B. R. Reddy, “Diabetes prediction using ensemble learning techniques,” Procedia Computer Science, vol. 258, pp. 3155–3164, 2025, doi: 10.1016/j.procs.2025.04.573.
W. Li, Y. Peng, and K. Peng, “Diabetes prediction model based on GA-XGBoost and stacking ensemble algorithm,” PLoS ONE, vol. 19, no. 9, p. e0311222, 2024, doi: 10.1371/journal.pone.0311222.
Z. Rafie, M. S. Talab, B. E. Z. Koor, A. Garavand, C. Salehnasab, and M. Ghaderzadeh, “Leveraging XGBoost and explainable AI for accurate prediction of type 2 diabetes,” BMC Public Health, vol. 25, no. 1, p. 3688, 2025, doi: 10.1186/s12889-025-24953-w.
K. Pang, “A comparative study of explainable machine learning models with Shapley values for diabetes prediction,” Healthcare Analytics, vol. 7, p. 100390, 2025, doi: 10.1016/j.health.2025.100390.
H. Almarzooq and U. bin Waheed, “Automating hyperparameter optimization in geophysics with Optuna: A comparative study,” Geophysical Prospecting, vol. 72, pp. 1778–1788, 2024, doi: 10.1111/1365-2478.13484.
A. M. Salih, Z. Raisi-Estabragh, I. B. Galazzo, P. Radeva, S. E. Petersen, K. Lekadir, and G. Menegaz, “A perspective on explainable artificial intelligence methods: SHAP and LIME,” Advanced Intelligent Systems, vol. 7, no. 1, Art. no. 2400304, 2025, doi: 10.1002/aisy.202400304.
P. Netayawijit, W. Chansanam, and K. Sorn-In, “Interpretable machine learning framework for diabetes prediction: Integrating SMOTE balancing with SHAP explainability for clinical decision support,” Healthcare, vol. 13, no. 20, p. 2588, 2025, doi: 10.3390/healthcare13202588.
V. Q. Tran, Y. Choi, and H. Byeon, “Explainable stacking ensemble with feature tokenizer transformers for men's diabetes prediction,” Journal of Men's Health, vol. 20, no. 11, Art. no. 38, 2024, doi: 10.22514/jomh.2024.184.
Md. M. Islam, H. R. Rifat, Md. S. B. Shahid, A. Akhter, M. A. Uddin, and K. M. M. Uddin, “Explainable machine learning for efficient diabetes prediction using hyperparameter tuning, SHAP analysis, partial dependency, and LIME,” Engineering Reports, vol. 7, no. 1, p. e13080, 2025, doi: 10.1002/eng2.13080.
M. Liu et al., “FAIM: Fairness-aware interpretable modeling for trustworthy machine learning in healthcare,” Patterns, vol. 5, no. 10, p. 101059, 2024, doi: 10.1016/j.patter.2024.101059.
V. N. Dang, V. M. Campello, J. Hernández-González, and K. Lekadir, “Empirical comparison of post-processing debiasing methods for machine learning classifiers in healthcare,” Journal of Healthcare Informatics Research, vol. 9, no. 3, pp. 465–493, 2025, doi: 10.1007/s41666-025-00196-7.
A. García-Galindo, M. López-De-Castro, and R. Armañanzas, “Fair prediction sets through multi-objective hyperparameter optimization,” Machine Learning, vol. 114, no. 1, Art. no. 27, 2025, doi: 10.1007/s10994-024-06721-w.
V. P. Nugraha and A. Febrian, “Comprehensive diabetes risk prediction using BRFSS data: Performance, explainability, fairness, and calibration,” Journal of Applied Informatics and Computing, vol. 10, no. 3, pp. 2357–2368, 2026, doi: 10.30871/jaic.v10i3.12740.
M. R. Sanjaya, A. Khoiriah, R. I. Heroza, and B. W. Putra, “Analysis of user reviews for the MyTelkomsel app using naïve Bayes and Random Forest methods,” Jurnal Testing dan Implementasi Sistem Informasi, vol. 4, no. 1, pp. 32–48, 2026, doi: 10.55583/jtisi.v4i1.2220.
M. H. H. Alkinani, A. M. Albugami, and M. R. Ibrahim, “Data preprocessing techniques in healthcare machine learning: A systematic review,” Healthcare Analytics, vol. 6, Art. no. 100336, 2024, doi: 10.1016/j.health.2024.100336.
G. R. Ashisha, X. A. Mary, E. G. M. Kanaga, J. Andrew, and R. J. Eunice, “Random oversampling-based diabetes classification via machine learning algorithms,” International Journal of Computational Intelligence Systems, vol. 17, no. 1, Art. no. 270, 2024, doi: 10.1007/s44196-024-00678-3.
B. W. Putra, “Deep neural network classification for ECG Arrhythmia Detection with Optimized Feature Autoencoder,” Journal of Advanced Research in Applied Mechanics, vol. 133, no. 1, pp. 78–86, 2025, doi: 10.37934/aram.133.1.7886.
M. Frasca, D. La Torre, G. Pravettoni, and I. Cutica, “Explainable and interpretable artificial intelligence in medicine: A systematic bibliometric review,” Discover Artificial Intelligence, vol. 4, no. 1, Art. no. 15, 2024, doi: 10.1007/s44163-024-00114-7.
J. Caterson, A. Lewin, and E. Williamson, “The application of explainable artificial intelligence (XAI) in electronic health record research: A scoping review,” Digital Health, vol. 10, Art. no. 20552076241272657, 2024, doi: 10.1177/20552076241272657.
C. Lokker, W. Abdelkader, E. Bagheri, et al., “Boosting efficiency in a clinical literature surveillance system with LightGBM,” PLOS Digital Health, vol. 3, no. 9, Art. no. e0000299, 2024, doi: 10.1371/journal.pdig.0000299.
Bila bermanfaat silahkan share artikel ini
Berikan Komentar Anda terhadap artikel Comparative Evaluation of Machine Learning Models with Explainability and Fairness for Diabetes Risk Prediction
Pages: 786-798
Copyright (c) 2026 Vanya Dwi Nabila, Bayu Wijaya Putra, M Rudi Sanjaya, Dwi Rosa Indah
This work is licensed under a Creative Commons Attribution 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 Creative Commons Attribution 4.0 International License that allows others to share the work with an acknowledgment 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 acknowledgment 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 (Refer to The Effect of Open Access).
