Evaluation of Tree-Based Models for Predicting Social Assistance Recipient Status Based on National Socio-Economic Survey (SUSENAS) 2024
Journal of Mathematics, Computations and Statistics · 03.03.2026 · 9 (1) · 15-38
DOI: 10.35580/xyyv0f37 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Mathematics, Computations and Statistics, online erschienen am 03.03.2026. Verfasst von Yani Prihantini Hiola und 7 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-SA 4.0).
Abstract (Originalsprache)
Poverty is a major socioeconomic challenge in Indonesia that affects the effectiveness of social protection programs. In response to this challenge, the government has created social assistance programs to improve the welfare of the people. However, the distribution of social assistance is often considered to be inaccurate, resulting in households that are deemed eligible for social assistance not being identified as recipients. One solution to improve the accuracy of distribution is the application of machine learning in the context of classification. Several tree-based models, such as LightGBM, Random Forest, and XGBoost, were selected because of their superior capabilities compared to classical models such as logistic regression, especially in handling complex data and fulfilling model assumptions. This study compares the performance of these three models in predicting social assistance recipient status using data from the 2024 West Java Provincial National Socioeconomic Survey (SUSENAS). Model evaluation was conducted on several data pre-processing scenarios involving outlier handling, class balancing, and feature engineering. The results show that LightGBM consistently outperforms the other models on six metrics, namely Accuracy, Balanced Accuracy, F1-Score, ROC-AUC, PR-AUC, and Brier Score, out of a total of eight evaluation metrics used. SHAP analysis identifies Social Assistance History and Asset Score as the most influential features for model prediction. Friedman and Nemenyi nonparametric tests confirmed significant performance differences between LightGBM and other models based on the F1-Score, PR-AUC, and Brier Score metrics. These findings indicate that tree-based models, particularly LightGBM, can support the development of a more targeted and data-driven social assistance targeting system. Keywords: Social Assistance; Tree-Based; SHAP; SUSENAS; Hybrid Bayesian Optimization
Zitieren
APA
Hiola, Y. P., Zulhijrah, Putra, I. G. N. S., Limba, S. Z., Sartono, B., Firdawanti, A. R., Susetyo, B., & Dito, G. A. (2026). Evaluation of Tree-Based Models for Predicting Social Assistance Recipient Status Based on National Socio-Economic Survey (SUSENAS) 2024. Journal of Mathematics, Computations and Statistics, 9(1), 15-38. https://doi.org/10.35580/xyyv0f37
Vancouver
Hiola YP, Zulhijrah, Putra IGNS, Limba SZ, Sartono B, Firdawanti AR, et al. Evaluation of Tree-Based Models for Predicting Social Assistance Recipient Status Based on National Socio-Economic Survey (SUSENAS) 2024. Journal of Mathematics, Computations and Statistics. 2026;9(1):15-38. doi:10.35580/xyyv0f37
BibTeX
@article{Hiola2026,
author = {Hiola, Yani Prihantini and Zulhijrah and Putra, I Gusti Ngurah Sentana and Limba, Syella Zignora and Sartono, Bagus and Firdawanti, Aulia Rizki and Susetyo, Budi and Dito, Gerry Alfa},
title = {Evaluation of Tree-Based Models for Predicting Social Assistance Recipient Status Based on National Socio-Economic Survey (SUSENAS) 2024},
journal = {Journal of Mathematics, Computations and Statistics},
year = {2026},
volume = {9},
number = {1},
pages = {15--38},
doi = {10.35580/xyyv0f37}
}
EUVIMED