Prediksi Risiko Diabetes Menggunakan Random Forest dengan Evaluasi Klasifikasi dan Analisis Fitur
Keywords:
diabetes, Random Forest, klasifikasi risiko, pembelajaran mesin, ROC-AUC, feature importanceAbstract
Prediksi risiko diabetes berbasis data merupakan salah satu penerapan pembelajaran mesin yang memerlukan evaluasi kinerja dan keterbatasan penggunaan secara jelas. Penelitian ini menyusun implementasi Random Forest untuk klasifikasi biner risiko diabetes sekaligus menganalisis kontribusi relatif atribut. Data yang digunakan berupa 768 record simulasi dengan delapan variabel numerik dan label dua kelas. Prosedur meliputi persiapan data, pembagian stratifikasi 80:20, pembentukan model dengan 180 pohon, evaluasi pada 154 data uji, serta repeated stratified five-fold cross-validation sebanyak dua pengulangan. Hasil pengujian menunjukkan accuracy 83,77%, precision 76,67%, recall 80,70%, F1-score 78,63%, dan ROC-AUC 87,39%. Pengujian silang menghasilkan rerata accuracy 80,33% dengan simpangan baku 3,17%, F1-score 73,06% dengan simpangan baku 3,82%, serta ROC-AUC 85,95% dengan simpangan baku 2,95%. Atribut Glucose memiliki nilai feature importance tertinggi sebesar 0,4495, diikuti BMI sebesar 0,1357 dan DiabetesPedigreeFunction sebesar 0,0876. Temuan ini menunjukkan bahwa alur pelatihan, evaluasi, dan penjelasan model dapat diverifikasi secara numerik. Namun, data berasal dari simulasi dan tidak membuktikan ketepatan klinis, generalisasi populasi, ataupun kelayakan penggunaan sebagai alat diagnosis. Pengujian pada data pasien terverifikasi, validasi eksternal, pemeriksaan kalibrasi, dan evaluasi risiko kesalahan diperlukan sebelum pemanfaatan di layanan kesehatan.
References
[1] H. Lai, H. Huang, K. Keshavjee, A. Guergachi, and X. Gao, “Predictive models for diabetes mellitus using machine learning techniques,” BMC Endocrine Disorders, vol. 19, art. 101, 2019, doi: 10.1186/s12902-019-0436-6.
[2] Z. Xie, O. Nikolayeva, J. Luo, and D. Li, “Building Risk Prediction Models for Type 2 Diabetes Using Machine Learning Techniques,” Preventing Chronic Disease, vol. 16, art. E130, 2019, doi: 10.5888/pcd16.190109.
[3] A. Dinh, S. Miertschin, A. Young, and S. D. Mohanty, “A data-driven approach to predicting diabetes and cardiovascular disease with machine learning,” BMC Medical Informatics and Decision Making, vol. 19, art. 211, 2019, doi: 10.1186/s12911-019-0918-5.
[4] A. Mujumdar and V. Vaidehi, “Diabetes Prediction using Machine Learning Algorithms,” Procedia Computer Science, vol. 165, pp. 292–299, 2019, doi: 10.1016/j.procs.2020.01.047.
[5] S. Kaul and Y. Kumar, “Artificial Intelligence-based Learning Techniques for Diabetes Prediction: Challenges and Systematic Review,” SN Computer Science, vol. 1, art. 322, 2020, doi: 10.1007/s42979-020-00337-2.
[6] L. J. Muhammad, E. A. Algehyne, and S. S. Usman, “Predictive Supervised Machine Learning Models for Diabetes Mellitus,” SN Computer Science, vol. 1, art. 240, 2020, doi: 10.1007/s42979-020-00250-8.
[7] U. M. Butt, S. Letchmunan, M. Ali, F. H. Hassan, A. Baqir, and H. H. R. Sherazi, “Machine Learning Based Diabetes Classification and Prediction for Healthcare Applications,” Journal of Healthcare Engineering, vol. 2021, art. 9930985, 2021, doi: 10.1155/2021/9930985.
[8] C. Olisah, O. Adeleye, L. Smith, and M. Smith, “A Robust Machine Learning Framework for Diabetes Prediction,” Future Technologies Conference 2021, vol. 359, pp. 775–792, 2021, doi: 10.1007/978-3-030-89880-9_58.
[9] V. Chang, M. A. Ganatra, K. Hall, L. Golightly, and Q. A. Xu, “An assessment of machine learning models and algorithms for early prediction and diagnosis of diabetes using health indicators,” Healthcare Analytics, vol. 2, art. 100118, 2022, doi: 10.1016/j.health.2022.100118.
[10] S. Kodama et al., “Predictive ability of current machine learning algorithms for type 2 diabetes mellitus: A meta-analysis,” Journal of Diabetes Investigation, vol. 13, no. 5, pp. 900–908, 2022, doi: 10.1111/jdi.13736.
[11] M. O. Olusanya, R. E. Ogunsakin, M. Ghai, and M. A. Adeleke, “Accuracy of Machine Learning Classification Models for the Prediction of Type 2 Diabetes Mellitus: A Systematic Survey and Meta-Analysis Approach,” IJERPH, vol. 19, no. 21, art. 14280, 2022, doi: 10.3390/ijerph192114280.
[12] J. B. Ginting, T. Suci, C. N. Ginting, and D. D. Hartono, “Early detection system of risk factors for diabetes mellitus type 2 utilization of machine learning-random forest,” Journal of Family and Community Medicine, vol. 30, no. 3, pp. 171–179, 2023, doi: 10.4103/jfcm.jfcm_33_23.
[13] Sriyanto and A. R. Supriyatna, “Prediksi Penyakit Diabetes Menggunakan Algoritma Random Forest,” TEKNIKA, vol. 17, no. 1, pp. 163–172, 2023, doi: 10.5281/zenodo.8051410.
[14] Andi, Thamrin, A. Susanto, E. Wijaya, and D. Djohan, “Analysis of the random forest and grid search algorithms in early detection of diabetes mellitus disease,” Jurnal Mantik, vol. 7, no. 2, 2023, doi: 10.35335/mantik.v7i2.3981.
[15] C. N. Noviyanti and A. Alamsyah, “Early Detection of Diabetes Using Random Forest Algorithm,” Journal of Information System Exploration and Research, vol. 2, no. 1, 2024, doi: 10.52465/joiser.v2i1.245.
[16] L. Breiman, “Random Forests,” Machine Learning, vol. 45, pp. 5–32, 2001, doi: 10.1023/A:1010933404324.
[17] G. S. Collins et al., “TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods,” BMJ, vol. 385, art. e078378, 2024, doi: 10.1136/bmj-2023-078378.
[18] B. Van Calster, D. J. McLernon, M. van Smeden, L. Wynants, and E. W. Steyerberg, “Calibration: the Achilles heel of predictive analytics,” BMC Medicine, vol. 17, art. 230, 2019, doi: 10.1186/s12916-019-1466-7.
[19] D. Chicco and G. Jurman, “The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation,” BMC Genomics, vol. 21, art. 6, 2020, doi: 10.1186/s12864-019-6413-7.
[20] S. M. Lundberg et al., “From local explanations to global understanding with explainable AI for trees,” Nature Machine Intelligence, vol. 2, pp. 56–67, 2020, doi: 10.1038/s42256-019-0138-9.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Berto Nadeak
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
