Abstract
Obesity classification models often rely heavily on anthropometric features, particularly weight and height, because both variables directly contribute to body mass index (BMI) calculation. This study investigates the extent of such dependency using feature ablation and error analysis. Experiments were conducted on the NObeyesdad dataset using Logistic Regression and Random Forest under three feature configurations: (1) baseline using all features, (2) ablation without weight, and (3) ablation without height. Model performance was evaluated using Balanced Accuracy, Macro F1-score, confusion matrices, and class-level analysis. Results show that Random Forest consistently outperformed Logistic Regression across all scenarios and demonstrated greater robustness to feature removal. Under the baseline setting, Random Forest achieved a Balanced Accuracy of 0.90 and a Macro F1-score of 0.90, while Logistic Regression obtained 0.89 for both metrics. Performance degradation became more pronounced when weight was removed compared with height, indicating stronger model dependence on weight-related information. Error analysis further revealed increased confusion among adjacent obesity categories, particularly overweight classes. These findings highlight the importance of feature dependency evaluation to improve robustness and interpretability in obesity classification models.
References
[1] S. K. Saraswati et al., “Literature Review : Faktor Risiko Penyebab Obesitas,” Media Kesehatan Masyarakat Indonesia, vol. 20, no. 1, pp. 70–74, 2021, doi: 10.14710/mkmi.20.1.70-74.
[2] B. Busebee, W. Ghusn, L. Cifuentes, and A. Acosta, “Obesity: A Review of Pathophysiology and Classification,” Mayo Clin. Proc., vol. 98, no. 12, pp. 1842–1857, 2024, doi: 10.1016/j.mayocp.2023.05.026.Obesity.
[3] T. Islam and M. S. Ali, “Ensemble Learning for Healthcare: A Comparative Analysis of Hybrid Voting and Ensemble Stacking in Obesity Risk Prediction,” pp. 1–26, 2025, [Online]. Available: http://arxiv.org/abs/2509.02826
[4] M. Andini, B. Budirman, and B. U. Hasanah, “Faktor Predisposisi, Klasifikasi Subgrup, Biomarker dan Mekanisme Pencegahan Obesitas : Tinjauan Pustaka,” Media Kesehatan Politeknik Kesehatan Makassar, vol. 20, no. 1, pp. 88–96, 2025, doi: 10.32382/medkes.v20i1.1414.
[5] S. A. Utiarahman and A. M. M. Pratama, “Penerapan Support Vector Machine dan Random Forest Classifier Untuk Klasifikasi Tingkat Obesitas,” Jurnal Fasilkom, vol. 14, no. 3, pp. 754–760, 2024, doi: 10.37859/jf.v14i3.8104.
[6] N. H. Phelps et al., “Worldwide trends in underweight and obesity from 1990 to 2022: a pooled analysis of 3663 population-representative studies with 222 million children, adolescents, and adults,” The Lancet, vol. 403, no. 10431, pp. 1027–1050, Mar. 2024, doi: 10.1016/S0140-6736(23)02750-2.
[7] F. R. Valerian, M. Syarief, and D. A. Fatah, “Klasifikasi Tingkat Obesitas Menggunakan Metode Gbm Dan Confusion Matrix,” JATI (Jurnal Mahasiswa Teknik Informatika), vol. 9, no. 2, pp. 2242–2249, 2025, doi: 10.36040/jati.v9i2.13062.
[8] A. K. Tripathi, N. R. Chauhan, and A. Sharma, “Obesity classification and prognosis using machine learning,” AIP Conf. Proc., vol. 3224, no. 1, p. 20050, Feb. 2025, doi: 10.1063/5.0245938.
[9] K. A. Wardana and A. M. A. Rahim, “Analisis Perbandingan Algoritma XGBoost Dan Algoritma Random Forest Untuk Klasifikasi Data Kesehatan Mental,” Jurnal Ilmu Komputer dan Pendidikan, vol. 2, no. 5, pp. 808–818, 2024, [Online]. Available: https://www.kaggle.com/datasets/bhavikjikadara/mental-health-dataset
[10] G. Kasneci and E. Kasneci, “Enriching Tabular Data with Contextual LLM Embeddings: A Comprehensive Ablation Study for Ensemble Classifiers,” 2024, [Online]. Available: http://arxiv.org/abs/2411.01645
[11] G. Elkhawaga, O. Elzeki, M. Abuelkheir, and M. Reichert, “Evaluating Explainable Artificial Intelligence Methods Based on Feature Elimination: A Functionality-Grounded Approach,” Electronics (Basel)., vol. 12, no. 7, p. 1670, Mar. 2023, doi: 10.3390/electronics12071670.
[12] S. Sheikholeslami, M. Meister, T. Wang, A. H. Payberah, V. Vlassov, and J. Dowling, “AutoAblation: Automated Parallel Ablation Studies for Deep Learning,” in Proceedings of the 1st Workshop on Machine Learning and Systems, New York, NY, USA: ACM, Apr. 2021, pp. 55–61. doi: 10.1145/3437984.3458834.
[13] C. Özkurt, “Examination and Evaluation of Obesity Risk Factors with Explainable Artificial Intelligence,” Computers and Electronics in Medicine, vol. 1, no. 1, pp. 12–17, 2024, doi: 10.69882/adba.cem.2024072.
[14] Z. Helforoush and H. Sayyad, “Prediction and classification of obesity risk based on a hybrid metaheuristic machine learning approach,” Front. Big Data, vol. 7, 2024, doi: 10.3389/fdata.2024.1469981.
[15] S. Y. Sibi and A. R. Widiarti, “Klasifikasi Tingkat Obesitas Mempergunakan Algoritma KNN,” in SEMINAR NASIONAL CORISINDO, Bali, 2022, pp. 370–375. [Online]. Available: https://corisindo.stikom-bali.ac.id/penelitian/index.php/semnas/article/view/73
[16] T. Khater, H. Tawfik, S. Sowdagar, and B. Singh, “Interpretable Models For ML-Based Classification of Obesity,” in Proceedings of the 2023 7th International Conference on Cloud and Big Data Computing, in ICCBDC ’23. New York, NY, USA: Association for Computing Machinery, 2023, pp. 40–47. doi: 10.1145/3616131.3616137.
[17] V. Çetİn, “Pamukkale Üniversitesi Mühendislik Bili mleri Dergisi Pamukkale University Journal of Engineering Sciences A comprehensive review on data preprocessing techniques in data analysis Veri analizinde veri ön işleme teknikleri üzerine kapsamlı bir inceleme,” vol. 28, no. 2, pp. 299–312, 2022, doi: 10.5505/pajes.2021.62687.
[18] A. Muniasamy, AI Model Design and Data Management for Disease Prediction. in Advances in Computational Intelligence and Robotics. IGI Global Scientific Publishing, 2025. doi: 10.4018/979-8-3373-5137-7.
[19] J. M. A. S. Dachi and P. Sitompul, “Comparison Analysis of the XGBoost Algorithm and Random Forest Ensemble Learning Algorithm in Credit Decision Classification,” Jurnal Riset Rumpun Matematika dan Ilmu Pengetahuan Alam, vol. 2, no. 2, pp. 87–103, 2023, doi: https://doi.org/10.55606/jurrimipa.v2i2.1336.
[20] A. Rahim, I. Pratiwi, and M. Fikri, “Klasifikasi Penyakit Jantung Menggunakan Metode Synthetic Minority Over-Sampling Technique dan Random Forest,” Indonesian Journal of Computer Science, vol. 12, no. 1, pp. 2995–3011, 2023.
[21] S. S. Al Malaky, A. A. C. Nisa, S. Armiyanti, and R. S. Setyawan, “Comparative Study of Obesity Levels Classification,” JEECS (Journal of Electrical Engineering and Computer Sciences), vol. 10, no. 1, pp. 69–75, 2025, doi: 10.54732/jeecs.v10i1.8.
[22] W. Elouataoui, I. El Alaoui, S. El Mendili, and Y. Gahi, “An End-to-End Big Data Deduplication Framework based on Online Continuous Learning,” International Journal of Advanced Computer Science and Applications, vol. 13, no. 9, pp. 281–291, 2022, doi: 10.14569/IJACSA.2022.0130933.
[23] F. Aldi et al., “Standardscaler ’ S Potential in Enhancing Breast Cancer,” Journal of Applied Engineering and Technological Science (JAETS), vol. 5, no. 1, pp. 401–413, 2023, doi: 10.37385/jaets.v5i1.3080.
[24] E. Poslavskaya and A. Korolev, “Encoding categorical data: Is there yet anything ‘hotter’ than one-hot encoding?,” pp. 92–107, 2023, [Online]. Available: http://arxiv.org/abs/2312.16930
[25] L. Yu, R. Zhou, R. Chen, and K. K. Lai, “Missing Data Preprocessing in Credit Classification: One-Hot Encoding or Imputation?,” Emerging Markets Finance and Trade, vol. 58, no. 2, pp. 472–482, Jan. 2022, doi: 10.1080/1540496X.2020.1825935.
[26] K. J. Olowe, N. L. Edoh, S. J. C. Zouo, and J. Olamijuwon, “Comprehensive review of logistic regression techniques in predicting health outcomes and trends,” World Journal of Advanced Pharmaceutical and Life Sciences, vol. 7, no. 2, pp. 016–026, 2024, doi: 10.53346/wjapls.2024.7.2.0039.
[27] A. M. Carreira-Perpiñán and S. S. Hada, “Inverse classification with logistic and softmax classifiers: efficient optimization,” Transactions on Machine Learning Research, vol. March, no. 2, pp. 1–21, Mar. 2026, [Online]. Available: http://arxiv.org/abs/2309.08945
[28] Y. Yao, J. Zou, and H. Wang, “Optimal Poisson Subsampling for Softmax Regression,” J. Syst. Sci. Complex., vol. 36, no. 4, pp. 1609–1625, 2023, doi: 10.1007/s11424-023-1179-z.
[29] H. A. Salman, A. Kalakech, and A. Steiti, “Random Forest Algorithm Overview,” Babylonian Journal of Machine Learning, vol. 2024, pp. 69–79, 2024, doi: 10.58496/BJML/2024/007.
[30] M. Bilal, G. Ali, M. W. Iqbal, M. Anwar, M. S. A. Malik, and R. A. Kadir, “Auto-Prep: Efficient and Automated Data Preprocessing Pipeline,” IEEE Access, vol. 10, no. October, pp. 107764–107784, 2022, doi: 10.1109/ACCESS.2022.3198662.
[31] M. Kumar and V. Bhardwaj, Evaluating Label Encoding and Preprocessing Techniques for Breast Cancer Prediction Using Machine Learning Algorithms, vol. 18, no. 1. Springer Netherlands, 2025. doi: 10.1007/s44196-025-00957-7.
[32] S. Sathyanarayanan, “Confusion Matrix-Based Performance Evaluation Metrics,” African Journal of Biomedical Research, vol. 27, no. 4, pp. 4023–4031, 2024, doi: 10.53555/ajbr.v27i4s.4345.
[33] A. Gevaert, A. Jan, R. Thijs, B. Dirk, and V. Tijl, Evaluating feature attribution methods in the image domain, vol. 113, no. 9. Springer US, 2024. doi: 10.1007/s10994-024-06550-x.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.