AI Research Paper Summarizer and Recommendation Engine

Abstract

In the modern research ecosystem, the volume of published academic papers has increased exponentially, making it extremely difficult for students, researchers, and professionals to keep up with relevant literature. Traditional methods of manually reading and filtering research papers are time-consuming and inefficient. This research proposes an intelligent system that automates research paper summarization and relevant paper recommendation using advanced Natural Language Processing techniques. The system leverages transformer-based models and FAISS (Facebook AI Similarity Search) for semantic summarization and similarity-based recommendation. Research papers are dynamically fetched using the arXiv API, ensuring access to the latest research content. The system architecture integrates MongoDB for scalable NoSQL storage, FastAPI for backend API handling, and Streamlit for frontend interaction. The proposed system significantly reduces cognitive load by delivering concise summaries, relevant recommendations, and real-time responses. The system improves research productivity, knowledge discovery, and academic workflow efficiency.