Year: 2025
Type: article
Abstract: In modern academic, legal, and business settings, users frequently encounter the problem of reading extensive documents in PDF format. This problem is resolved by DocuSense AI, which powers a sophisticated web application with Generative AI, allowing users to engage with their documents using natural language queries. Built with Python, Lang Chain, and Llama 3.3, the application transforms static documents into interactive and intelligent dialogues. The system generates vector embeddings of the PDFs for quick querying, while the frontend interface is user-friendly and consists of HTML, CSS, and JavaScript. PostgreSQL is employed for secure username and password registration. DocuSense AI is tailored for students, scholars, and working professionals to enhance their comprehension and productivity. Literature reviews support the implementation of Generative AI in conjunction with semantic search and layout-aware models in the analysis of documents centred around human users. This solution changes the way documents are engaged with—making the process quicker, smarter, and more efficient. Keywords-Generative AI, Llama 3.3, Lang Chain, PDF Querying, Natural Language Processing, Document Summarization, AI Chatbot, Document Interaction.
DOI: 10.55041/ijsrem52986
Authors: Jyoti Madake, Kritika Raina, Akanksha Kamble
Institution: Vishwakarma Institute of Technology
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4415077718
FWCI:
Cites:
Cited by:
Related to: