Meningkatkan Klasifikasi Obesitas Multi-Kelas Menggunakan Hybrid Stacking dan Meta-Learner CatBoost yang Interpretable melalui Analisis SHAP Level-2

Septiani Wulandari Lomi; Slamet Sudaryanto

doi:10.47065/bits.v7i3.8802

Septiani Wulandari Lomi Universitas Dian Nuswantoro, Semarang, Indonesia
Slamet Sudaryanto * Universitas Dian Nuswantoro, Semarang, Indonesia

(*) Corresponding Author

DOI: https://doi.org/10.47065/bits.v7i3.8802

Keywords: Hybrid Stacking; CatBoost; Obesity Prediction; Multiclass; SHAP Level-2

Abstract

Obesity is a global health problem that requires accurate, stable, and transparent multi-class prediction methods to support early clinical intervention. Previous studies used a Hybrid Stacking architecture with a linear Meta-Learner, which achieved 96.88% accuracy but had limitations in capturing complex non-linear interactions between basic model predictions. The main problem lies in the limitations of the linear Meta-Learner (Logistic Regression), which is not optimal in integrating non-linear signals from tree-based models at Level-1. The purpose of this study is to improve the performance, stability, and transparency of multi-class obesity predictions through the development of a Hybrid Stacking architecture with a non-linear Meta-Learner and the implementation of model interpretability techniques. To address this gap, this study proposes a new Hybrid Stacking Ensemble model by replacing the linear Meta-Learner with a powerful boosting model, namely CatBoost. The proposed model was evaluated on a multi-class obesity dataset and successfully surpassed state-of-the-art (SOTA) performance. The main performance improvement is demonstrated by an increase in Accuracy to 97.83% (an absolute increase of +0.95%) and a significant improvement in multi-class stability metrics: MCC (reaching 97.30%) and Cohen's Kappa (reaching 97.39%). This superiority validates the hypothesis that non-linear Meta-Learners are more effective. Furthermore, we included the technical innovation of Manual Padding on Level-1 outputs to ensure feature consistency, enabling a valid SHAP Level-2 analysis. The SHAP analysis revealed a strategic synergy, where CatBoost relied on Logistic Regression (a linear model) to predict high-risk class probabilities (Obesity Type II & III), while utilizing tree-based models for other classes. This model provides a superior, stable, and transparent methodology for obesity level prediction.

Downloads

Download data is not yet available.

References

Z. Yao, B. G. Tchang, M. Albert, R. S. Blumenthal, K. Nasir, and M. J. Blaha, “Associations between Class I, II, or III Obesity and Health Outcomes,” NEJM Evidence, vol. 4, no. 4, Mar. 2025, doi: 10.1056/EVIDoa2400229.

R. Kaur, R. Kumar, and M. Gupta, “Predicting risk of obesity and meal planning to reduce the obese in adulthood using artificial intelligence,” Endocrine, vol. 78, no. 3, pp. 458–469, Oct. 2022, doi: 10.1007/s12020-022-03215-4.

N. Khodadadi, M. Saber, and M. Abotaleb, “A Data-Driven Approach for Obesity Classification using Machine Learning,” Journal of Artificial Intelligence and Metaheuristics, vol. 3, no. 2, pp. 08–17, 2023, doi: 10.54216/JAIM.030201.

H. Lim, H. Lee, and J. Kim, “A prediction model for childhood obesity risk using the machine learning method: a panel study on Korean children,” Sci Rep, vol. 13, no. 1, p. 10122, Jun. 2023, doi: 10.1038/s41598-023-37171-4.

S. Jeon, M. Kim, J. Yoon, S. Lee, and S. Youm, “Machine learning-based obesity classification considering 3D body scanner measurements,” Sci Rep, vol. 13, no. 1, p. 3299, Feb. 2023, doi: 10.1038/s41598-023-30434-0.

W. Lin, S. Shi, H. Huang, J. Wen, and G. Chen, “Predicting risk of obesity in overweight adults using interpretable machine learning algorithms,” Front Endocrinol (Lausanne), vol. 14, Nov. 2023, doi: 10.3389/fendo.2023.1292167.

U. Sarmah, P. Borah, and D. K. Bhattacharyya, “Ensemble Learning Methods: An Empirical Study,” SN Comput Sci, vol. 5, no. 7, p. 924, Oct. 2024, doi: 10.1007/s42979-024-03252-y.

D. D. Solomon et al., “Hybrid Majority Voting: Prediction and Classification Model for Obesity,” Diagnostics, vol. 13, no. 15, p. 2610, Aug. 2023, doi: 10.3390/diagnostics13152610.

I. D. Mienye and Y. Sun, “A Survey of Ensemble Learning: Concepts, Algorithms, Applications, and Prospects,” 2022, Institute of Electrical and Electronics Engineers Inc. doi: 10.1109/ACCESS.2022.3207287.

S. M. Ganie, P. K. D. Pramanik, and Z. Zhao, “Lifestyle data-based multiclass obesity prediction with interpretable ensemble models incorporating SHAP and LIME analysis,” Sci Rep, vol. 15, no. 1, p. 36916, Oct. 2025, doi: 10.1038/s41598-025-20936-4.

I. G. S. M. Diayasa, M. Idhom, A. Fauzi, and A. T. Damaliana, “Stacking Ensemble Methods to Predict Obesity Levels in Adults,” in 2022 IEEE 8th Information Technology International Seminar (ITIS), IEEE, Oct. 2022, pp. 339–344. doi: 10.1109/ITIS57155.2022.10010260.

E. Ileberi and Y. Sun, “A Hybrid Deep Learning Ensemble Model for Credit Card Fraud Detection,” IEEE Access, vol. 12, pp. 175829–175838, 2024, doi: 10.1109/ACCESS.2024.3502542.

Y. Li, G. Liu, Y. Cao, J. Chen, X. Gang, and J. Tang, “WNPS-LSTM-Informer: A Hybrid Stacking model for medium-term photovoltaic power forecasting with ranked feature selection,” Renew Energy, vol. 244, p. 122687, May 2025, doi: 10.1016/j.renene.2025.122687.

A. Ferreras et al., “Systematic Review of Machine Learning applied to the Prediction of Obesity and Overweight,” J Med Syst, vol. 47, no. 1, p. 8, Jan. 2023, doi: 10.1007/s10916-022-01904-1.

T. Khater, H. Tawfik, and B. Singh, “Machine Learning for the Classification of Obesity Levels Based on Lifestyle Factors,” in Proceedings of the 2023 7th International Conference on Cloud and Big Data Computing, New York, NY, USA: ACM, Aug. 2023, pp. 28–33. doi: 10.1145/3616131.3616135.

I. D. Mienye, Y. Sun, and Z. Wang, “An improved ensemble learning approach for the prediction of heart disease risk,” Inform Med Unlocked, vol. 20, p. 100402, 2020, doi: 10.1016/j.imu.2020.100402.

V. Hassija et al., “Interpreting Black-Box Models: A Review on Explainable Artificial Intelligence,” Jan. 01, 2024, Springer. doi: 10.1007/s12559-023-10179-8.

W. Khan et al., “Predicting preterm birth using explainable machine learning in a prospective cohort of nulliparous and multiparous pregnant women,” PLoS One, vol. 18, no. 12, p. e0293925, Dec. 2023, doi: 10.1371/journal.pone.0293925.

M. Nauta et al., “From Anecdotal Evidence to Quantitative Evaluation Methods: A Systematic Review on Evaluating Explainable AI,” ACM Comput Surv, vol. 55, no. 13s, pp. 1–42, Dec. 2023, doi: 10.1145/3583558.

K. Zhuang, C. Zhang, Z. Chen, T. She, and M. Wang, “Integrating convolutional neural networks with ensemble methods for enhanced diabetes diagnosis: a multi-dataset evaluation,” Front Med (Lausanne), vol. 12, Sep. 2025, doi: 10.3389/fmed.2025.1657889.

M. Nagassou, R. W. Mwangi, and E. Nyarige, “A Hybrid Ensemble Learning Approach Utilizing Light Gradient Boosting Machine and Category Boosting Model for Lifestyle-Based Prediction of Type-II Diabetes Mellitus,” Journal of Data Analysis and Information Processing, vol. 11, no. 04, pp. 480–511, 2023, doi: 10.4236/jdaip.2023.114025.

Bila bermanfaat silahkan share artikel ini

Berikan Komentar Anda terhadap artikel Meningkatkan Klasifikasi Obesitas Multi-Kelas Menggunakan Hybrid Stacking dan Meta-Learner CatBoost yang Interpretable melalui Analisis SHAP Level-2