AI-Driven Welfare Scheme Recommendation Using Random Forest and RAG

We propose a scalable framework that integrates a multi-output Random Forest classifier with a retrieval-augmented generation module to recommend government welfare schemes to citizens based on their demographic profiles. The system first applies a preprocessing pipeline that normalizes raw input features—such as age, income, occupation, and caste—using a fuzzy matching algorithm to resolve lexical inconsistencies in categorical variables. A multi-label Random Forest ensemble, comprising hundreds of decision trees, then predicts eligibility probabilities across all available schemes simultaneously, and a calibrated confidence threshold selects a candidate subset of schemes. To ensure factual accuracy in the natural language explanations delivered to users, we incorporate a retrieval-augmented generation component. This module embeds verified scheme descriptions into a high-dimensional vector space, retrieves the most relevant document chunks for the candidate schemes using cosine similarity, and feeds both the retrieved context and the user’s original query into an instruction-tuned large language model. The classification stage thus acts as a computational filter that narrows the retrieval search space, thereby improving both system efficiency and response precision. The primary contribution of this work lies in the novel coupling of an ensemble-based eligibility predictor with a retrieval-constrained generative model, which prevents hallucinated outputs while remaining adaptable to large-scale, heterogeneous citizen data. Experimental evaluations on synthetic datasets, designed to mimic real-world public records, demonstrate that the framework achieves high precision in eligibility prediction and generates coherent, evidence-backed recommendations. This approach has significant implications for making complex social welfare systems more accessible to underserved populations.

Artificial Intelligence, Government Schemes, Eligibility Prediction, Machine Learning, Random Forest, Decision Tree, RAG Chatbot,

1. C Malhotra (2024) Digital India: Past, present and future. Indien im 21. Jahrhundert: Auf dem Weg zur digitalen Großmacht. [Google Scholar] [Crossref]

2. M Kos & L Foreman (2001) Using expert systems to deliver better public services. Canberra Bulletin of Public Administration. [Google Scholar] [Crossref]

3. H Linusson (2013) Multi-output random forests. diva-portal.org. [Google Scholar] [Crossref]

4. P Lewis, E Perez, A Piktus, F Petroni, et al. (2020) Retrieval-augmented generation for knowledge-intensive NLP tasks. Advances in Neural Information Processing Systems. [Google Scholar] [Crossref]

5. S Chaudhuri, K Ganjam, V Ganti, et al. (2003) Robust and efficient fuzzy match for online data cleaning. Proceedings of. [Google Scholar] [Crossref]

6. N Reimers & I Gurevych (2019) Sentence-BERT: Sentence embeddings using siamese BERT-networks. Proceedings of. [Google Scholar] [Crossref]

7. MSM AL-Inizi (2025) Enhancing governmental decision-making through predictive analytics with machine learning-based data-driven framework. Babylonian Journal of Machine Learning. [Google Scholar] [Crossref]

8. G Tsoumakas & I Katakis (2007) Multi-label classification: An overview. International Journal of Data Warehousing and Mining. [Google Scholar] [Crossref]

9. W Zhang & J Zhang (2025) Hallucination mitigation for retrieval-augmented large language models: a review. Mathematics. [Google Scholar] [Crossref]

10. F Neha, D Bhati & DK Shukla (2025) Retrieval-augmented generation (RAG) in healthcare: A comprehensive review. AI. [Google Scholar] [Crossref]

11. N Wiratunga, R Abeyratne, L Jayawardena, et al. (2024) CBR-RAG: case-based reasoning for retrieval augmented generation in LLMs for legal question answering. International Conference on Case-Based Reasoning. [Google Scholar] [Crossref]

12. A Alsubayhay & M Abdalla (2024) Enhancing citizen engagement in E-government services through AI-driven chatbots. Sebha University Conference Proceedings. [Google Scholar] [Crossref]

13. L Wang, M Tan & J Han (2016) FastHybrid: A hybrid model for efficient answer selection. Proceedings of COLING. [Google Scholar] [Crossref]

14. G Hui & MR Hayllar (2010) Creating public value in e-Government: A public-private-citizen collaboration framework in Web 2.0. Australian Journal of Public Administration. [Google Scholar] [Crossref]

15. V Leelavathi & S Pavithra (2026) AI-Based Prediction of Beneficiary Eligibility for Government Welfare Schemes. International Journal of Engineering and Technical Research. [Google Scholar] [Crossref]

16. L Breiman (2001) Random forests. Machine Learning. [Google Scholar] [Crossref]

17. X Wu, Y Gao & D Jiao (2019) Multi-label classification based on random forest algorithm for non-intrusive load monitoring system. Processes. [Google Scholar] [Crossref]

18. H Touvron, L Martin, K Stone, P Albert, et al. (2023) Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288. [Google Scholar] [Crossref]

19. M Douze, A Guzhva, C Deng, J Johnson, et al. (2025) The FAISS library. IEEE Transactions on Big Data. [Google Scholar] [Crossref]

20. V Unnikrishnan & KS Imai (2020) Does the old-age pension scheme improve household welfare? Evidence from India. World Development. [Google Scholar] [Crossref]

21. International Institute for Population Sciences (IIPS), et al. (2021) National Family Health Survey (NFHS-5), India, 2019–21. International Institute for Population Sciences. [Google Scholar] [Crossref]

22. A Sengupta (2023) Pandemic on employment and earning in urban India during the first three months of pandemic period: An analysis with unit-level data of periodic labour force survey. The Indian Journal of Labour Economics. [Google Scholar] [Crossref]

23. L von Puttkamer (2016) India: Slum-free by 2022? A people-centered evaluation of the Pradhan Mantri Awas Yojana Scheme. ETH Zurich. [Google Scholar] [Crossref]

24. J Read & F Perez-Cruz (2014) Deep learning for multi-label classification. arXiv preprint arXiv:1502.05988. [Google Scholar] [Crossref]

25. T Zhang, V Kishore, F Wu, KQ Weinberger, et al. (2019) BERTScore: Evaluating text generation with BERT. arXiv preprint arXiv:1904.09675. [Google Scholar] [Crossref]

26. F Pedregosa, G Varoquaux, A Gramfort, et al. (2011) Scikit-learn: Machine learning in Python. Journal of Machine Learning Research. [Google Scholar] [Crossref]

27. M Veale, M Van Kleek & R Binns (2018) Fairness and accountability design needs for algorithmic support in high-stakes public sector decision-making. [Google Scholar] [Crossref]

28. C Dwork, M Hardt, T Pitassi, O Reingold, et al. (2012) Fairness through awareness. Proceedings of the 3rd Innovations in Theoretical Computer Science Conference. [Google Scholar] [Crossref]

29. L Loezer, F Enembreck, JP Barddal, et al. (2020) Cost-sensitive learning for imbalanced data streams. Proceedings of the 35th Annual ACM Symposium on Applied Computing. [Google Scholar] [Crossref]

30. G Hovakimyan & JM Bravo (2024) Evolving strategies in machine learning: a systematic review of concept drift detection. Information. [Google Scholar] [Crossref]

31. HM Gomes, A Bifet, J Read, JP Barddal, F Enembreck, et al. (2017) Adaptive random forests for evolving data stream classification. Machine Learning. [Google Scholar] [Crossref]

32. MK Chan (2013) A dynamic model of welfare reform. Econometrica. [Google Scholar] [Crossref]

33. N Mehdiyev, C Houy, O Gutermuth, L Mayer, et al. (2021) Explainable artificial intelligence (XAI) supporting public administration processes – on the potential of XAI in tax audit processes. International Conference on Electronic Government. [Google Scholar] [Crossref]

34. SM Lundberg & SI Lee (2017) A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems. [Google Scholar] [Crossref]

35. S Kumar, S Datta, V Singh, D Datta, SK Singh, et al. (2024) Applications, challenges, and future directions of human-in-the-loop learning. IEEE Access. [Google Scholar] [Crossref]

• Block-Based Programming for Education: A Comprehensive Analysis of Visual Programming Environments in K-12 Learning
• Management of Academic Libraries and Client Satisfaction Towards Digital Utilization: Basis for Monitoring Library Operations in SOCCSKSARGEN Region.
• Revenue Leakages in TPA Insurance Claims and Corporate Claims: An Institutional Overview of Aster Prime Hospital, Hyderabad
• Technology and Innovation in Hospitality and Tourism: A Management Perspective
• Geospatial Distribution of Tarok Sacred Grove of Langtang North and Langtang South Local Government Areas