Explainable Data-Driven Machine Learning for Identifying MBG Program Beneficiaries in Medan City

Solly Solly Aryza, Al Khowarizmi, Muhammad Furqon, Zulkarnain Lubis, Abdul Razak Nasution

Abstract


The effectiveness of social assistance programs depends heavily on the accuracy and transparency of beneficiary identification. In many urban areas, including Medan, challenges such as incomplete data, administrative bias, and inefficient targeting often lead to inclusion and exclusion errors in determining beneficiaries of the MBG (Makan Bergizi Gratis) program. This study aims to develop an explainable data-driven machine learning model to improve the accuracy and transparency of identifying eligible MBG program beneficiaries. The research employs a quantitative approach using socio-economic and demographic datasets collected from local government records, including variables such as household income, employment status, education level, household size, housing conditions, and access to public services. Several machine learning algorithms, including Random Forest, Gradient Boosting, and Logistic Regression, are implemented to classify potential beneficiaries. To enhance transparency and interpretability, the model integrates Explainable Artificial Intelligence (XAI) techniques, such as SHAP (Shapley Additive Explanations), to identify the most influential factors affecting eligibility predictions. The results demonstrate that the proposed data-driven model significantly improves the accuracy of beneficiary classification while providing interpretable insights into key socio-economic indicators influencing eligibility. The findings indicate that income level, employment status, household dependency ratio, and housing conditions are among the most critical determinants in identifying eligible recipients. The implementation of explainable machine learning models supports more transparent and accountable decision-making in social assistance programs. This research contributes to the development of data-driven governance by providing a robust analytical framework for improving the targeting efficiency of social welfare programs in urban areas. Practically, the proposed framework can assist policymakers and local government agencies in designing fairer and more efficient beneficiary identification systems for the MBG program in Medan City, ultimately supporting better resource allocation and improved social welfare outcomes.

Keywords


Data-driven machine learning; Explainable AI (XAI); MBG program beneficiaries; social assistance targeting; urban data analytics; Medan.

Full Text:

PDF

References


Agrawal, R., & Sharma, S. (2024). Significance of explainable artificial intelligence models in decision making. Journal of Artificial Intelligence Research, 81, 211–229. https://doi.org/10.1613/jair.1.14845

Aryza S et al, (2024). A ROBUST OPTIMIZATION TO DYNAMIC SUPPLIER DECISIONS AND SUPPLY ALLOCATION PROBLEMS IN THE MULTI-RETAIL INDUSTRY. Eastern-European Journal of Enterprise Technologies, 129(3).

Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324

Goodfellow, I., Bengio, Y., & Courville, A. (2022). Deep learning and artificial intelligence: Foundations and applications. MIT Press.

James, G., Witten, D., Hastie, T., & Tibshirani, R. (2021). An introduction to statistical learning with applications in R(2nd ed.). Springer. https://doi.org/10.1007/978-1-0716-1418-1

Kumar, V., Singh, R., & Patel, A. (2023). Machine learning approaches for social welfare targeting in developing countries. Applied Artificial Intelligence, 37(1), 217–235. https://doi.org/10.1080/08839514.2023.2189120

Molnar, C. (2022). Interpretable machine learning: A guide for making black box models explainable (2nd ed.). Lulu Press.

OECD. (2021). The path to becoming a data-driven public sector. Organisation for Economic Co-operation and Development. https://doi.org/10.1787/059814a7-en

Rudin, C. (2022). Interpretable machine learning: Fundamental principles and challenges. Nature Machine Intelligence, 4(5), 373–381. https://doi.org/10.1038/s42256-021-00410-w

Roscher, R., Bohn, B., Duarte, M., & Garcke, J. (2021). Explainable machine learning for scientific insights and discoveries. IEEE Access, 8, 42200–42216. https://doi.org/10.1109/ACCESS.2020.2976199

Shapley, L. S. (2021). A value for n-person games and explainable AI foundations. Game Theory and Machine Learning Review, 12(3), 145–162.

United Nations. (2022). Data strategy of the United Nations for action by everyone, everywhere. United Nations Publications.

World Bank. (2021). Data-driven development: Using big data for social protection and poverty reduction. World Bank Publications.

Zhang, Y., Li, H., & Wang, J. (2022). Using explainable artificial intelligence to identify key characteristics of deep poverty. Sustainability, 14(15), 9123. https://doi.org/10.3390/su14159123

Zhao, L., Chen, X., & Huang, J. (2023). Machine learning applications in social policy analysis: A systematic review. Government Information Quarterly, 40(3), 101789. https://doi.org/10.1016/j.giq.2023.101789




DOI: https://doi.org/10.30596/jcositte.v7i1.29560

Refbacks

  • There are currently no refbacks.