AI Powered College Enquiry Chatbot

Abstract

Educational institutions receive numerous enquiries from students and parents regarding admissions, courses, fees, examinations, and placements. Handling these enquiries manually increases the workload of administrative staff and may lead to delayed or inconsistent responses. To overcome this problem, this project presents an AI-Powered College Enquiry Chatbot that automates common college-related queries using Retrieval-Augmented Generation (RAG).The system is developed using a three-tier architecture consisting of a React frontend, Node.js/Express.js backend, and MongoDB database. The chatbot uses LangChain and Google Generative AI to understand user queries and generate relevant responses. Frequently Asked Questions (FAQs) are converted into vector embeddings and stored in a vector store. When a user submits a query, the system performs similarity search to identify the most relevant FAQ. If the similarity score satisfies a predefined confidence threshold, the retrieved information is provided to the language model to generate a suitable response. If no relevant information is found, the chatbot provides a safe fallback response, reducing the possibility of incorrect or hallucinated answers.The system also provides an administrative dashboard for managing FAQ content and monitoring chatbot performance. JWT authentication and bcrypt password hashing are used to secure administrator access. User conversations and feedback are stored in MongoDB along with useful metadata such as category, confidence score, and fallback status. This helps administrators identify missing information and improve the chatbot\'s knowledge base. The proposed chatbot provides fast, consistent, and reliable responses while reducing the workload of college administrative staff. The project demonstrates the practical use of RAG and conversational AI in the educational domain. Future enhancements include multilingual support, persistent vector storage, advanced analytics, and cloud deployment.

Introduction

The text describes the design and development of an AI-Powered College Enquiry Chatbot intended to automate and improve communication between educational institutions and students, parents, and other users. Traditional enquiry methods such as office visits, phone calls, emails, and websites can require significant human effort and may not provide immediate responses. The proposed chatbot addresses these limitations by providing 24/7 conversational access to college information.

The system uses Artificial Intelligence (AI), Natural Language Processing (NLP), Retrieval-Augmented Generation (RAG), Large Language Models (LLMs), text embeddings, and vector similarity search to understand questions and provide relevant answers. It can handle enquiries concerning admissions, courses, fees, eligibility, examinations, placements, and college facilities.

Literature Survey

Previous educational chatbots have generally relied on predefined FAQs, rules, or keyword matching. While these systems can answer common questions quickly, they may struggle when users phrase questions differently or ask questions outside their predefined datasets. NLP-based systems improve understanding by identifying the intent and meaning of queries, while modern RAG-based systems improve reliability by retrieving relevant information from an institutional knowledge base before generating a response.

The proposed system combines these approaches with a web interface, database, administrator dashboard, conversation history, and feedback mechanism, allowing college information to be updated as requirements change.

Major Challenges

The project identifies several important challenges:

• Understanding different natural-language formulations of the same question.

• Keeping information such as fees, admission dates, and examination schedules accurate and updated.

• Handling questions for which the knowledge base has no suitable answer.

• Integrating the frontend, backend, AI/RAG services, and database effectively.

• Protecting administrator accounts and conversation data through authentication and authorization.

• Achieving accurate responses while keeping response time low.

• Maintaining a simple and user-friendly interface.

Proposed Methodology

The development follows a modular and iterative approach. First, the requirements and major enquiry categories are identified. A knowledge base is then created containing FAQs and college information.

The backend is developed with APIs for authentication, FAQ management, chatbot communication, feedback, and analytics. MongoDB is used to store FAQs, administrator information, conversations, and feedback.

When a user submits a question, the system processes it using NLP, searches the college knowledge base for relevant information, and passes the retrieved content to an AI model. The model then generates a contextual response, which is returned to the web interface.

The complete system is tested through functional, integration, and performance testing to evaluate accuracy, reliability, response time, and usability.

Algorithms and Techniques

The major technologies used are:

• NLP: Understands user questions and identifies their intent.

• RAG: Retrieves relevant college information before generating an answer.

• Text Embeddings: Convert queries and stored information into numerical vectors.

• Vector Similarity Search: Finds information based on semantic meaning rather than exact keywords.

• LLMs: Generate natural and context-aware responses.

• CRUD Operations: Allow administrators to create, read, update, and delete information.

• Authentication and Authorization: Protect administrative functions.

• Conversation Logging and Feedback: Help monitor interactions and improve the knowledge base.

System Architecture

The architecture consists of four main layers:

Frontend → Backend → AI/RAG Services → MongoDB Database

The frontend provides the chat interface. The backend manages APIs, authentication, FAQs, conversations, feedback, and communication with AI services. The AI/RAG layer interprets queries, retrieves relevant information, and generates responses. MongoDB stores the required institutional and interaction data. An administrator interface allows authorized users to update the knowledge base.

Conclusion

The AI-Powered College Enquiry Chatbot provides an efficient and intelligent solution for handling common college-related enquiries. The system uses Artificial Intelligence, Natural Language Processing, and Retrieval-Augmented Generation to understand user queries and provide relevant information about admissions, courses, fees, examinations, placements, and other college activities. The chatbot reduces the repetitive workload of administrative staff and allows users to access information quickly through a simple conversational interface. The admin dashboard enables authorized users to manage FAQs, update information, monitor conversations, and analyze chatbot performance. Overall, the proposed system improves the accessibility, efficiency, and user experience of college enquiry services and provides a strong foundation for future enhancements such as multilingual support, voice interaction, and integration with other college management systems.

References

[1] A. Vaswani et al., “Attention is all you need,” in Proc. 31st Int. Conf. Neural Information Processing Systems (NIPS), Long Beach, CA, USA, 2017, pp. 6000–6010. [2] P. Lewis et al., “Retrieval-augmented generation for knowledge-intensive NLP tasks,” in Proc. 34th Int. Conf. Neural Information Processing Systems (NeurIPS), 2020, pp. 9459–9474. [3] J. Weizenbaum, “ELIZA — a computer program for the study of natural language communication between man and machine,” Commun. ACM, vol. 9, no. 1, pp. 36–45, Jan. 1966. [4] T. Mikolov, K. Chen, G. Corrado, and J. Dean, “Efficient estimation of word representations in vector space,” in Proc. Int. Conf. Learning Representations (ICLR) Workshop, 2013. [5] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova, “BERT: Pre-training of deep bidirectional transformers for language understanding,” in Proc. NAACL-HLT, Minneapolis, MN, USA, 2019, pp. 4171–4186. [6] LangChain, “LangChain documentation,” LangChain Inc. [Online]. Available: https://docs.langchain.com. Accessed: Aug. 2026. [7] Google, “Google AI for Developers — Gemini API documentation,” Google LLC. [Online]. Available: https://ai.google.dev. Accessed: Aug. 2026. [8] MongoDB Inc., “MongoDB manual,” MongoDB, Inc. [Online]. Available: https://www.mongodb.com/docs/manual/. Accessed: Aug. 2026. [9] OpenJS Foundation, “Express.js — Node.js web application framework,” [Online]. Available: https://expressjs.com. Accessed: Aug. 2026. [10] Meta Platforms, Inc., “React — the library for web and native user interfaces,” [Online]. Available: https://react.dev. Accessed: Aug. 2026.

Copyright

Copyright © 2026 G. Vamsi, N. Charan. 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.