University Admission Helpdesk

This AI-powered university chatbot uses an LSTM neural network and Flask to answer common admission and campus queries. Featuring an interactive web interface, it provides instant, intent-based responses and intelligently routes low-confidence questions to human administrators, ensuring a seamless user experience.

The University Admission Helpdesk Chatbot is an intelligent, end-to-end web application designed to streamline student inquiries regarding campus life and admissions. Built using Python, Flask, and an LSTM (Long Short-Term Memory) neural network via TensorFlow, the application processes natural language queries to instantly answer common questions about available courses, fee structures, department heads, and travel guidance. By training a custom classifier on specific university intents, the system maps user text to the most relevant category and serves accurate, real-time responses through a clean, interactive browser chat interface. To ensure reliability and a positive user experience, the chatbot features a smart fallback mechanism; if the model's prediction confidence drops below a specific threshold, it gracefully defers the conversation and guides the user to contact a human administrator. This project demonstrates a complete machine learning pipeline, from natural language preprocessing and model training to deploying a lightweight, accessible web frontend.

See more: Link