RAG-BASED LEGAL/GOVERNMENT SCHEME ASSISTANT

Access to accurate information about government welfare schemes remains a major challenge for many citizens, particularly in rural and underserved communities. Although official portals provide scheme details, users often struggle to identify programs relevant to their eligibility due to complex documentation and scattered data sources. This paper presents SchemeAssist, an intelligent AI-powered chatbot built using Retrieval-Augmented Generation (RAG) to deliver reliable, context-aware responses regarding government schemes.The proposed system integrates a large language model with a vector database containing curated government scheme documents. When a user submits a query, the system retrieves relevant information and generates precise responses grounded in verified data, thereby reducing hallucinations commonly associated with generative AI models. The chatbot supports natural language interaction, enabling users to ask eligibility questions, required documents, benefits, and application procedures.Experimental evaluation demonstrates improved response accuracy, reduced misinformation, and enhanced user accessibility compared to traditional