International Journal of Scientific Research in Engineering and Management (IJSREM)
Volume: 09 Issue: 02 | Feb - 2025 SJIF Rating: 8.448 ISSN: 2582-3930
AI bot that interacts with multiple pdf's
Pratham V1, Likhith HJ2
1Dept of Artificial Intelligence and Data Science, East West institute of technology
2Dept of Artificial Intelligence and Data Science, East West institute of technology
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - This project involves the development of an
intelligent AI bot capable of interacting with and extracting
information from multiple PDF documents. Leveraging 3. CONCLUSIONS
advanced natural language processing (NLP) techniques, the
AI bot can read, comprehend, and respond to user queries based The "Chat with PDFs" project demonstrates a significant
on the content of multiple PDF files. The bot ensures efficient advancement in document analysis by combining AI-driven
data retrieval, providing concise and accurate answers, conversational interfaces with robust semantic search
summaries, or insights. This tool aims to streamline the process mechanisms. It simplifies information retrieval, reduces
of document management and information extraction, making manual effort, and showcases the potential of emerging AI
it an invaluable asset for research, business, and educational technologies.
purposes.
ACKNOWLEDGEMENT
Key Words: AI Bot, NLP, PDF Interaction, Information This project was a collaborative effort made possible by the
Extraction, Document Management, Data Retrieval, Multi- hard work and dedication of its authors, Pratham V and Likhith
PDF Analysis, User Queries, Automated Summarization, HJ. We would like to extend our deepest gratitude to Akshatha
Research Tool, Business Intelligence, Educational Tool. K for their invaluable mentorship and support throughout the
development of this project. Their guidance and insights have
been instrumental in shaping our research and outcomes.
We would also like to thank our colleagues and friends for their
1.INTRODUCTION constant support, constructive feedback, and encouragement.
Their contributions have been essential in refining our work
In today's digital age, information retrieval from extensive and achieving our goals.
documents can be tedious and time-consuming. "Chat with Our heartfelt thanks go to our families for their unwavering
PDFs" is an innovative solution designed to enhance the user support and understanding during this journey. Their belief in
experience by enabling interaction with PDF documents via a us has been a constant source of motivation.
conversational interface. This application leverages advanced Lastly, we acknowledge the use of various resources and tools,
natural language processing (NLP) techniques, powered by including advanced AI and NLP technologies, which have
Google Generative AI and LangChain, to enable users to ask empowered the development of this project.
questions and receive relevant answers directly from PDF Pratham V & Likhith HJ
content..
REFERENCES
2. Body of Paper 1.LangChainDocumentation:
[https://langchain.com](https://langchain.com)
Efficiently extracting relevant information from extensive PDFs 2.FAISSDocumentation:
is challenging due to the lack of interactive and context-aware [https://github.com/facebookresearch/faiss](https://github.com/fac
tools. Basic PDF viewers provide search functionality but are ebookresearch/faiss)
limited to exact keyword matches. Time-consuming for users to 3.PyPDF2Library:
sift through large amounts of irrelevant data. A Python-based [https://pypdf2.readthedocs.io](https://pypdf2.readthedocs.io)
solution using **Streamlit** for a user-friendly interface 4. Streamlit Framework: [https://streamlit.io](https://streamlit.io)
5.Google Generative AI: [https://cloud.google.com/generative-
Leverages **Google Generative AI** for contextual
ai](https://cloud.google.com/generative-ai)
understanding of PDF content. Implements **FAISS** for
efficient semantic search using embeddings generated by
LangChain. Enable conversational interaction with PDF
documents. Facilitate multi-document semantic search.
Improve user productivity by reducing the time spent searching.
© 2025, IJSREM | www.ijsrem.com DOI: 10.55041/IJSREM41277 | Page 1
