CampusGPT - Your Campus AI Assistant

Abstract

With the proliferation of digital academic resources, it is becoming increasingly challenging for a student to find a specific piece of information within a large academic document. The existing search mechanisms in academic documents rely heavily on matching keywords and do not consider the actual meaning of the search query. This is making the process of information retrieval inefficient and time-consuming. In this paper, an academic assistant tool called CampusGPT is proposed. The proposed tool is based on an artificial intelligence algorithm that will enable the student to interact with the document in a natural language. The proposed academic assistant tool is based on a combination of semantic search and a generative artificial intelligence algorithm. The proposed academic assistant tool will process the uploaded document in the PDF format and then convert it into a vector. The proposed academic assistant tool is implemented using the latest technologies. The proposed academic assistant tool will improve the efficiency and speed of the document-based query mechanism.

Introduction

CampusGPT is an AI-based academic assistant designed to improve information retrieval from educational documents. Traditional keyword-based search systems are often inefficient because they require exact word matches, making it difficult and time-consuming for users to find relevant information.

The proposed system uses Natural Language Processing and semantic search to allow users to ask questions in a conversational manner and receive accurate, context-based answers. It integrates document processing, embedding-based retrieval, and AI-generated responses into a single platform.

The system follows a layered architecture that includes user interface, document processing, embedding and retrieval, AI processing, and data management. It processes uploaded PDFs by extracting and splitting text, converting it into vector embeddings, and storing it in a database. When a user submits a query, the system retrieves the most relevant content using similarity search and generates a response using a language model.

CampusGPT provides fast response times (around 2–4 seconds), high accuracy, and improved user experience compared to traditional methods. It supports both document-based queries and general AI queries, making it flexible and scalable.

However, the system has some limitations, such as support only for PDF files and dependence on internet connectivity for AI-based responses. Overall, it significantly enhances academic information access by combining semantic search with AI-driven response generation.

Conclusion

The CampusGPT system demonstrates an effective solution for interacting with academic documents using natural language queries. By combining document processing, semantic search, and AI-based response generation, the system simplifies the process of retrieving relevant information. The results show that the system provides accurate and context-aware responses within a short time. The use of embedding-based retrieval enhances search quality compared to traditional keyword-based methods. The system also offers flexibility through its dual-mode functionality, allowing both document-based and general AI queries. However, the system currently supports only PDF documents and depends on internet connectivity for AI-based processing. Future work can focus on supporting additional file formats and improving system scalability. Overall, CampusGPT highlights the practical use of artificial intelligence in academic environments and its potential to improve learning experiences.

References

[1] T. Brown et al., “Language Models are Few-Shot Learners,” NeurIPS, 2020. [2] PyPDF2 Documentation, Python PDF Processing Library. [3] N. Reimers and I. Gurevych, “Sentence-BERT: Sentence Embeddings using Siamese BERT,” 2019. [4] J. Johnson et al., “FAISS: A Library for Efficient Similarity Search,” Facebook AI Research, 2017. [5] P. Lewis et al., “Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks,” 2020. [6] MySQL Documentation, Oracle Corporation. [7] LangChain Documentation, 2023. [8] A. Vaswani et al., “Attention Is All You Need,” NeurIPS, 2017.

Copyright

Copyright © 2026 Rachana Amale, Prof. G. P. Dhomse, Radhika Sharma, Trupti Gunjal, Siddhi Raundal. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.