Diyabet Tahmini için Makine Öğrenmesi Modellerinin Meta-Sezgisel Hiperparametre Optimizasyonu

Abstract

Bu çalışmada, makine öğrenmesine dayalı diyabet tahmininde meta-sezgisel hiperparametre optimizasyonunun etkisi ortak bir deneysel çerçevede incelenmiştir. Lojistik Regresyon, K-En Yakın Komşu, Karar Ağacı, Rastgele Orman ve XGBoost modelleri; Dhole Optimization Algorithm (DOA), Grey Wolf Optimization (GWO), Particle Swarm Optimization (PSO) ve RIME yöntemleri ile optimize edilmiştir. Tekrarlanan kayıtların çıkarılmasının ardından 96.146 gözlem kullanılmış; veri seti tabakalı biçimde eğitim ve bağımsız test kümelerine ayrılmıştır. Hiperparametre optimizasyonu yalnızca eğitim verisi üzerinde beş katlı çapraz doğrulama ile gerçekleştirilmiştir. Modeller doğruluk, kesinlik, duyarlılık, F1-skoru, Matthews Korelasyon Katsayısı, ROC-AUC ve PR-AUC ölçütleri ile değerlendirilmiştir. Toplam 20 model-optimizasyon birleşimi arasında XGBoost-PSO, 0,8119 F1-skoru ile en yüksek test performansını elde ederken, XGBoost-GWO 0,9781 ROC-AUC ve 0,8849 PR-AUC ile en yüksek sıralama performansını göstermiştir. Optimizasyon yöntemleri genelinde GWO en yüksek ortalama tahmin performansını sağlarken, DOA en yüksek hesaplama verimliliğini sunmuştur. Çapraz doğrulama ve bağımsız test sonuçlarının genel olarak uyumlu olduğu görülmüş; korelasyon ve permütasyon özellik önemi analizlerinde HbA1c düzeyi ile kan şekeri düzeyi en etkili değişkenler olarak belirlenmiştir. Sonuçlar, diyabet tahmininde model ve optimizasyon yöntemi seçiminin tahmin performansı, genellenebilirlik, yorumlanabilirlik ve hesaplama verimliliği birlikte dikkate alınarak yapılması gerektiğini göstermektedir.

Keywords

• diyabet tahmini
• makine öğrenmesi
• hiperparametre optimizasyonu
• meta-sezgisel optimizasyon
• XGBoost
• yorumlanabilirlik

Metaheuristic Hyperparameter Optimization of Machine Learning Models for Diabetes Prediction

Abstract

This study investigates the effect of metaheuristic hyperparameter optimization on machine learning-based diabetes prediction using a unified experimental framework. Five classifiers—Logistic Regression, K-Nearest Neighbors, Decision Tree, Random Forest, and XGBoost—were optimized using the Dhole Optimization Algorithm (DOA), Grey Wolf Optimization (GWO), Particle Swarm Optimization (PSO), and RIME. After removing duplicate records, 96,146 observations were retained and divided into stratified training and independent test subsets. Hyperparameter tuning was conducted only on the training data using five-fold cross-validation, while model performance was assessed using accuracy, precision, recall, F1 score, Matthews correlation coefficient, ROC-AUC, and PR-AUC. Among the 20 model-optimizer combinations, XGBoost-PSO achieved the highest test F1 score of 0.8119, whereas XGBoost-GWO produced the highest ROC-AUC of 0.9781 and PR-AUC of 0.8849. Decision Tree also showed competitive performance, while Random Forest-RIME achieved the highest recall but with lower precision and F1 score. At the optimizer level, GWO provided the strongest average predictive performance, whereas DOA offered the greatest computational efficiency. Cross-validation and independent test results were generally consistent, indicating satisfactory generalization. Correlation and permutation feature importance analyses identified HbA1c level and blood glucose level as the most influential predictors. Overall, the findings show that model and optimizer selection should jointly consider predictive performance, generalization, interpretability, and computational efficiency.

Keywords

• diabetes prediction
• machine learning
• hyperparameter optimization
• metaheuristic optimization
• XGBoost
• interpretability

References

1. [1] P. B. Khokhar, C. Gravino, and F. Palomba, “Advances in artificial intelligence for diabetes prediction: insights from a systematic literature review,” Artif. Intell. Med., vol. 164, p. 103132, Jun. 2025, doi: 10.1016/j.artmed.2025.103132.
2. [2] V. Jaiswal, A. Negi, and T. Pal, “A review on current advances in machine learning based diabetes prediction,” Prim. Care Diabetes, vol. 15, no. 3, pp. 435–443, Jun. 2021, doi: 10.1016/j.pcd.2021.02.005.
3. [3] D. Kalla, N. Smith, F. Samaah, and K. Polimetla, “Enhancing Early Diagnosis: Machine Learning Applications in Diabetes Prediction,” Journal of Artificial Intelligence & Cloud Computing, vol. 1, no. 1, p. 1, Jan. 2022, doi: 10.47363/JAICC/2022(1)191.
4. [4] E. M. Hameed, H. Joshi, and Q. K. Kadhim, “Advancements in Artificial Intelligence Techniques for Diabetes Prediction: A Comprehensive Literature Review,” Journal of Robotics and Control (JRC), vol. 6, no. 1, pp. 345–365, Feb. 2025, doi: 10.18196/jrc.v6i1.22258.
5. [5] M. E. Febrian, F. X. Ferdinan, G. P. Sendani, K. M. Suryanigrum, and R. Yunanda, “Diabetes prediction using supervised machine learning,” Procedia Comput. Sci., vol. 216, pp. 21–30, 2023, doi: 10.1016/j.procs.2022.12.107.
6. [6] J. J. Khanam and S. Y. Foo, “A comparison of machine learning algorithms for diabetes prediction,” ICT Express, vol. 7, no. 4, pp. 432–439, Dec. 2021, doi: 10.1016/j.icte.2021.02.004.
7. [7] R. Rastogi and M. Bansal, “Diabetes prediction model using data mining techniques,” Measurement: Sensors, vol. 25, p. 100605, Feb. 2023, doi: 10.1016/j.measen.2022.100605.
8. [8] G. Parimala, R. Kayalvizhi, and S. Nithiya, “Diabetes Prediction using Machine Learning,” in 2023 International Conference on Computer Communication and Informatics (ICCCI), IEEE, Jan. 2023, pp. 1–10. doi: 10.1109/ICCCI56745.2023.10128216.

9. [9] H. Zhou, Y. Xin, and S. Li, “A diabetes prediction model based on Boruta feature selection and ensemble learning,” BMC Bioinformatics, vol. 24, no. 1, p. 224, Jun. 2023, doi: 10.1186/s12859-023-05300-5.
10. [10] I. Tasin, T. U. Nabil, S. Islam, and R. Khan, “Diabetes prediction using machine learning and explainable AI techniques,” Healthc. Technol. Lett., vol. 10, no. 1–2, pp. 1–10, Feb. 2023, doi: 10.1049/htl2.12039.
11. [11] M. M. Bukhari, B. F. Alkhamees, S. Hussain, A. Gumaei, A. Assiri, and S. S. Ullah, “An Improved Artificial Neural Network Model for Effective Diabetes Prediction,” Complexity, vol. 2021, no. 1, Jan. 2021, doi: 10.1155/2021/5525271.
12. [12] A. Singh, A. Dhillon, N. Kumar, M. S. Hossain, G. Muhammad, and M. Kumar, “eDiaPredict: An Ensemble-based Framework for Diabetes Prediction,” ACM Transactions on Multimedia Computing, Communications, and Applications, vol. 17, no. 2s, pp. 1–26, Jun. 2021, doi: 10.1145/3415155.
13. [13] S. M. Ganie, P. K. D. Pramanik, M. Bashir Malik, S. Mallik, and H. Qin, “An ensemble learning approach for diabetes prediction using boosting techniques,” Front. Genet., vol. 14, Oct. 2023, doi: 10.3389/fgene.2023.1252159.
14. [14] M. O. Edeh et al., “A Classification Algorithm-Based Hybrid Diabetes Prediction Model,” Front. Public Health, vol. 10, Mar. 2022, doi: 10.3389/fpubh.2022.829519.
15. [15] A. E.-S. El-Bashbishy and H. M. El-Bakry, “Pediatric diabetes prediction using deep learning,” Sci. Rep., vol. 14, no. 1, p. 4206, Feb. 2024, doi: 10.1038/s41598-024-51438-4.
16. [16] J. Kaliappan et al., “Analyzing classification and feature selection strategies for diabetes prediction across diverse diabetes datasets,” Front. Artif. Intell., vol. 7, Aug. 2024, doi: 10.3389/frai.2024.1421751.
17. [17] N. Ahmed et al., “Machine learning based diabetes prediction and development of smart web application,” International Journal of Cognitive Computing in Engineering, vol. 2, pp. 229–241, Jun. 2021, doi: 10.1016/j.ijcce.2021.12.001.
18. [18] J. Ramesh, R. Aburukba, and A. Sagahyroon, “A remote healthcare monitoring framework for diabetes prediction using machine learning,” Healthc. Technol. Lett., vol. 8, no. 3, pp. 45–57, Jun. 2021, doi: 10.1049/htl2.12010.
19. [19] G. R. Kumar, R. V. Reddy, J. M, N. Pughazendi, S. Vidyullatha, and P. C. S. Reddy, “Web application based Diabetes prediction using Machine Learning,” in 2023 International Conference on Advances in Computing, Communication and Applied Informatics (ACCAI), IEEE, May 2023, pp. 1–7. doi: 10.1109/ACCAI58221.2023.10200323.
20. [20] S. Sivaranjani, S. Ananya, J. Aravinth, and R. Karthika, “Diabetes Prediction using Machine Learning Algorithms with Feature Selection and Dimensionality Reduction,” in 2021 7th International Conference on Advanced Computing and Communication Systems (ICACCS), IEEE, Mar. 2021, pp. 141–146. doi: 10.1109/ICACCS51430.2021.9441935.
21. [21] H. Gupta, H. Varshney, T. K. Sharma, N. Pachauri, and O. P. Verma, “Comparative performance analysis of quantum machine learning with deep learning for diabetes prediction,” Complex & Intelligent Systems, vol. 8, no. 4, pp. 3073–3087, Aug. 2022, doi: 10.1007/s40747-021-00398-7.
22. [22] H. El Massari, Z. Sabouri, S. Mhammedi, and N. Gherabi, “Diabetes Prediction Using Machine Learning Algorithms and Ontology,” Journal of ICT Standardization, May 2022, doi: 10.13052/jicts2245-800X.10212.
23. [23] D. B. Olawade, R. C. Owhonda, J. O. Alabi, E. Egbon, R. I. Ayo Daniel, and O. J. Bello, “Digital twin paradigm in diabetes prediction and management,” Diabetes Res. Clin. Pract., vol. 231, p. 113075, Jan. 2026, doi: 10.1016/j.diabres.2025.113075.
24. [24] J. Abdollahi and B. Nouri-Moghaddam, “Hybrid stacked ensemble combined with genetic algorithms for diabetes prediction,” Iran Journal of Computer Science, vol. 5, no. 3, pp. 205–220, Sep. 2022, doi: 10.1007/s42044-022-00100-1.
25. [25] S. Nahzat and M. Yağanoğlu, “Makine Öğrenimi Sınıflandırma Algoritmalarını Kullanarak Diyabet Tahmini,” European Journal of Science and Technology, Apr. 2021, doi: 10.31590/ejosat.899716.
26. [26] S. A. Alex, N. Jhanjhi, M. Humayun, A. O. Ibrahim, and A. W. Abulfaraj, “Deep LSTM Model for Diabetes Prediction with Class Balancing by SMOTE,” Electronics (Basel)., vol. 11, no. 17, p. 2737, Aug. 2022, doi: 10.3390/electronics11172737.
27. [27] H. Shao, X. Liu, D. Zong, and Q. Song, “Optimization of diabetes prediction methods based on combinatorial balancing algorithm,” Nutr. Diabetes, vol. 14, no. 1, p. 63, Aug. 2024, doi: 10.1038/s41387-024-00324-z.
28. [28] M. Salem Alzboon, M. Alqaraleh, and M. Subhi Al-Batah, “Diabetes Prediction and Management Using Machine Learning Approaches,” Data and Metadata, vol. 4, p. 545, Feb. 2025, doi: 10.56294/dm2025545.
29. [29] P. Cıhan and H. Coşkun, “Performance Comparison of Machine Learning Models for Diabetes Prediction,” in 2021 29th Signal Processing and Communications Applications Conference (SIU), IEEE, Jun. 2021, pp. 1–4. doi: 10.1109/SIU53274.2021.9477824.
30. [30] H. Rashid Abdulqadir, A. Mohsin Abdulazeez, and D. Assad Zebari, “Data Mining Classification Techniques for Diabetes Prediction,” Qubahan Academic Journal, vol. 1, no. 2, pp. 125–133, May 2021, doi: 10.48161/qaj.v1n2a55.
31. [31] M. Jichkar, R. Shende, O. Bonde, P. Agrawal, G. K. Gupta, and A. K. Singh, “Diabetes Prediction Using Machine Learning,” in 2024 IEEE Silchar Subsection Conference (SILCON 2024), IEEE, Nov. 2024, pp. 1–6. doi: 10.1109/SILCON63976.2024.10910769.
32. [32] J. Shin et al., “Development of Various Diabetes Prediction Models Using Machine Learning Techniques,” Diabetes Metab. J., vol. 46, no. 4, pp. 650–657, Jul. 2022, doi: 10.4093/dmj.2021.0115.
33. [33] M. Saleh Al Reshan et al., “An Innovative Ensemble Deep Learning Clinical Decision Support System for Diabetes Prediction,” IEEE Access, vol. 12, pp. 106193–106210, 2024, doi: 10.1109/ACCESS.2024.3436641.
34. [34] N. K. Trivedi, V. Gautam, H. Sharma, A. Anand, and S. Agarwal, “Diabetes Prediction using Different Machine Learning Techniques,” in 2022 2nd International Conference on Advance Computing and Innovative Technologies in Engineering (ICACITE), IEEE, Apr. 2022, pp. 2173–2177. doi: 10.1109/ICACITE53722.2022.9823640.
35. [35] M. Mustafa, “Diabetes Prediction Dataset,” Kaggle. Accessed: Aug. 09, 2025. [Online]. Available: https://www.kaggle.com/datasets/iammustafatz/diabetes-prediction-dataset/data
36. [36] S. Mirjalili, S. M. Mirjalili, and A. Lewis, “Grey Wolf Optimizer,” Advances in Engineering Software, vol. 69, pp. 46–61, Mar. 2014, doi: 10.1016/j.advengsoft.2013.12.007.
37. [37] F. Marini and B. Walczak, “Particle swarm optimization (PSO). A tutorial,” Chemometrics and Intelligent Laboratory Systems, vol. 149, pp. 153–165, Dec. 2015, doi: 10.1016/j.chemolab.2015.08.020.
38. [38] B. O. Mohammed, H. S. Aghdasi, and P. Salehpour, “Dhole optimization algorithm: a new metaheuristic algorithm for solving optimization problems,” Cluster Comput., vol. 28, no. 7, p. 430, Sep. 2025, doi: 10.1007/s10586-024-05005-1.
39. [39] H. Su et al., “RIME: A physics-based optimization,” Neurocomputing, vol. 532, pp. 183–214, May 2023, doi: 10.1016/j.neucom.2023.02.010.