AI Chatbot Project — Blossom Oyem

Rule-based and neural chatbot demo with Streamlit deployment and improved intent handling for teens.

Project Overview

This AI Chatbot project was developed as a practical teaching tool for teens aged 12–18. It combines a simple rule-based chatbot with an optional neural network model to improve natural language understanding. The project was deployed using Streamlit, allowing users to test the chatbot in a friendly, web-based interface.

Image Gallery

Screenshots of the AI Chatbot interface:

Key Features

Rule-based chatbot with pattern matching
Neural intent classifier using bag-of-words or embeddings
Streamlit GUI for easy testing
Teens-friendly simplified chatbot logic
Expandable intents and responses

Technologies Used

Python NLTK TensorFlow / PyTorch Streamlit Rule-based NLP

Development Workflow

Intent Design: Define intents, patterns, and responses for the chatbot.
Rule-based Matching: Extract keywords and map to predefined responses.
Neural Model (Optional): Build a simple classifier to detect intents.
Streamlit Deployment: Design textbox interface for interactive chats.
Testing & Refinement: Improve responses based on student feedback.

Sample Python Snippet

import random

responses = {
    "greeting": ["Hello!", "Hi there!", "Nice to meet you!"],
    "bye": ["Goodbye!", "See you later!", "Bye for now!"]
}

def get_intent(message):
    message = message.lower()
    if "hi" in message or "hello" in message:
        return "greeting"
    elif "bye" in message:
        return "bye"
    return "unknown"

def chatbot_reply(message):
    intent = get_intent(message)
    if intent in responses:
        return random.choice(responses[intent])
    return "I'm not sure I understand. Can you rephrase?"

Project Outcome

This project helped teens understand the fundamentals of chatbots, NLP, and intent classification. With a simple Streamlit deployment, students were able to interact with the chatbot and extend its capabilities.