Government subsidy programs play a crucial role in socio-economic development by supporting vulnerable populations in sectors such as agriculture, education, healthcare, energy, and food security. However, traditional subsidy management systems are often plagued by inefficiencies, fraud, leakage, lack of transparency, and poor targeting. The advent of digital governance and data-driven technologies has opened new avenues for reforming subsidy allocation and monitoring mechanisms. Machine learning (ML), in particular, offers powerful tools for automating eligibility assessment, predicting beneficiary behavior, detecting anomalies, and optimizing policy outcomes. This review paper presents a comprehensive analysis of online subsidy management systems integrated with machine learning techniques, with a specific focus on Logistic Regression, Decision Tree, and Random Forest algorithms. The paper discusses system architecture, data sources, preprocessing methods, algorithmic frameworks, evaluation metrics, real-world use cases, challenges, ethical considerations, and future research directions. The review aims to serve as a ready reference for researchers, policymakers, and system designers working toward intelligent, transparent, and efficient subsidy management platforms.
Biradar et al. (Thu,) studied this question.