AUTONOMOUS RESEARCH ASSISTANT

This paper presents an Autonomous Research Assistant based on Retrieval-Augmented Generation (RAG) to provide intelligent, accurate, and real-time information retrieval for users. The system is designed to collect data from multiple sources, including web content and structured knowledge bases, enabling precise and context-aware responses. Users can interact with the system through a simple web interface by providing textual queries. When a query is submitted, the system converts it into vector embeddings and retrieves the most relevant information using a vector database powered by FAISS. The retrieved content is then processed using a large language model such as Gemini to generate a coherent and human-like response. The system also maintains a database to store user queries, responses, and logs for efficient data management and future analysis. This approach significantly improves the accuracy and reliability of generated responses by grounding them in real-time retrieved data, reducing hallucination commonly observed in standalone language models. The proposed system demonstrates effective performance and provides a scalable solution for intelligent research assistance across various domains.