DOCQUERY BOT: OFFLINE DOCUMENT INTELLIGENCE VIA RAG WITH LOCAL LLMS

Shaik Mansoor; Shaik MuhammadMuzeer; Medha Yesu Babu; Tokka Kalyan; Shaik Mastanbi

DOCQUERY BOT: OFFLINE DOCUMENT INTELLIGENCE VIA RAG WITH LOCAL LLMS

April 2025

Vol-11, Issue-2

Paper ID: 26121

ISSN: 2395-4396

Downloads: 0

Abstract & Details

Research Area

Artificial Intelligence, Natural Language Processing (NLP), Information Retrieval, Local Large Language Models (LLMs), Document Intelligence

Keywords

RAG LLM GPT-4 LangChain Streamlit FAISS Local Chatbot PDF Query System Vector Similarity Privacy-Preserving AI

Abstract

Retrieval-Augmented Generation (RAG) systems combine the best of both retrieval-based and generative methods in natural language processing. This paper presents “DocQuery Bot,” a lightweight, privacy-focused RAG-based chatbot that operates entirely on local resources. Unlike cloud-dependent systems, our implementation ensures data privacy and offline functionality, using LangChain, OpenAI’s GPT-4, Streamlit, and FAISS for vector storage and similarity search. The chatbot is designed to provide domain-specific responses by retrieving chunks of information from user-uploaded PDF documents. The interface, powered by Streamlit, offers an intuitive chat experience for users, while the backend ensures quick and accurate retrieval with real-time response generation. This system is ideal for academic, legal, and private environments where cloud interaction is restricted or undesired. This paper outlines the design, architecture, implementation, evaluation metrics, and future enhancements for localized AI chatbot systems

License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Author Information

#	Name	Institute / Affiliation
1	Shaik Mansoor	Vasireddy Venkatadri Institute of Technology (VVIT)
2	Shaik MuhammadMuzeer	Vasireddy Venkatadri Institute of Technology (VVIT)
3	Medha Yesu Babu	Vasireddy Venkatadri Institute of Technology (VVIT)
4	Tokka Kalyan	Vasireddy Venkatadri Institute of Technology (VVIT)
5	Shaik Mastanbi	Vasireddy Venkatadri Institute of Technology (VVIT)

How to Cite

Use the following formats to cite this article in your research.

APA Style

Mansoor, Shaik, MuhammadMuzeer, Shaik, Babu, Medha Yesu, Kalyan, Tokka, & Mastanbi, Shaik (2025). DOCQUERY BOT: OFFLINE DOCUMENT INTELLIGENCE VIA RAG WITH LOCAL LLMS. International Journal of Advance Research and Innovative Ideas In Education, 11(2), 1379-1382.

MLA Style

Mansoor, Shaik, et al. "DOCQUERY BOT: OFFLINE DOCUMENT INTELLIGENCE VIA RAG WITH LOCAL LLMS." International Journal of Advance Research and Innovative Ideas In Education, vol. 11, no. 2, 2025, pp. 1379-1382.

IEEE Style

Shaik Mansoor, Shaik MuhammadMuzeer, Medha Yesu Babu, Tokka Kalyan, and Shaik Mastanbi, "DOCQUERY BOT: OFFLINE DOCUMENT INTELLIGENCE VIA RAG WITH LOCAL LLMS," International Journal of Advance Research and Innovative Ideas In Education, vol. 11, no. 2, pp. 1379-1382, 2025.

Vancouver Style

Mansoor Shaik, MuhammadMuzeer Shaik, Babu Medha Yesu, Kalyan Tokka, Mastanbi Shaik. DOCQUERY BOT: OFFLINE DOCUMENT INTELLIGENCE VIA RAG WITH LOCAL LLMS. International Journal of Advance Research and Innovative Ideas In Education. 2025;11(2):1379-1382.

Harvard Style

Mansoor, Shaik, MuhammadMuzeer, Shaik, Babu, Medha Yesu, Kalyan, Tokka, & Mastanbi, Shaik (2025) 'DOCQUERY BOT: OFFLINE DOCUMENT INTELLIGENCE VIA RAG WITH LOCAL LLMS', International Journal of Advance Research and Innovative Ideas In Education, 11(2), pp. 1379-1382.

Chicago Style

Mansoor, Shaik, et al. "DOCQUERY BOT: OFFLINE DOCUMENT INTELLIGENCE VIA RAG WITH LOCAL LLMS." International Journal of Advance Research and Innovative Ideas In Education 11, no. 2 (2025): 1379-1382.

Turabian Style

Mansoor, Shaik, et al. "DOCQUERY BOT: OFFLINE DOCUMENT INTELLIGENCE VIA RAG WITH LOCAL LLMS." International Journal of Advance Research and Innovative Ideas In Education 11, no. 2 (2025): 1379-1382.