Case Study
University of Abuja Chatbot
A chatbot that uses NLP and Naive Bayes intent classification to respond to common university queries.
Problem
Students often need quick answers to repeated questions about university processes, but information can be slow to access.
Solution
I built a chatbot that classifies user intent and provides relevant responses to common academic and administrative questions.
How It Works
Step 1
Users type a question into the chatbot interface.
Step 2
The input is preprocessed using NLP techniques.
Step 3
A Naive Bayes model predicts the user’s intent.
Step 4
The chatbot returns a matching response.
Challenges
Improving intent classification accuracy with limited training data.
Designing responses that feel useful and relevant.
Balancing simple ML methods with practical chatbot performance.
Outcome / Impact
The chatbot demonstrates applied machine learning for real-world support and information access.