Intelligent Document Query Hub — A Holistic Solution for Efficient PDF Information Retrieval and Extraction

Navigating large PDF documents to extract specific information is increasingly challenging due to the rapid growth of digital content. This paper presents an Intelligent PDF Question Answering (QA) system that ensures data privacy through the local deployment of the LLaMA 3.1 Multihop-RAG model using Ollama, orchestrated with LangChain. The architecture features a federated context catalog that integrates domain-specific resources, such as university academic guidelines, educational textbooks, clinical guidelines, financial compliance documents, and regulatory standards. The system combines dense retrieval with multi-hop reasoning and advanced context fusion to generate accurate, coherent responses across diverse document types. A joint optimization strategy enhances robustness and adaptability for complex queries. Document content is preprocessed and stored in a vector store, enabling fast, similarity-based retrieval. The privacy-preserving Smart PDF Document Query Hub using Multihop RAG achieves 85 % accuracy and an average response time of 30 seconds, making it a valuable tool for industry, researchers, students, and professionals seeking efficient information access.