CUTM CHATBOT - An AI-Powered University Assistant

A smart chatbot implementation designed for Centurion University of Technology and Management (CUTM). This project serves as an intelligent virtual assistant to help students, faculty, and visitors with university-related queries. Built using PyTorch and deployed with Flask and JavaScript, this chatbot demonstrates the practical application of AI in enhancing university communication systems.

About CUTM CHATBOT

CUTM CHATBOT is a college project developed for Centurion University of Technology and Management. It's designed to:

• Assist with common university-related queries
• Provide information about courses and programs
• Help with administrative questions
• Guide visitors through university resources
• Support students with general inquiries

Features

• Custom neural network implementation using PyTorch
• Natural Language Processing with NLTK
• Multiple deployment options:
  • Integrated Flask application with Jinja2 templates
  • Standalone REST API with separate frontend
• Real-time chat interface with JavaScript
• Customizable intents and responses specific to CUTM
• Easy-to-modify architecture

Deployment Options

1. Full Stack Flask Application

• Complete integration with Flask
• Server-side rendering using Jinja2 templates
• All-in-one solution

2. Decoupled Architecture

• Separate frontend and backend
• Flask serves only as REST API
• Frontend can be integrated into any application
• Greater flexibility and scalability

Prerequisites

• Python 3.7+
• Basic knowledge of Python, Flask, and JavaScript
• Understanding of REST APIs
• Familiarity with PyTorch (optional)

Installation

1. Clone the repository:

git clone https://github.com/maheshmm7/CUTM-Chatbot.git
cd CUTM-Chatbot

2. Create and activate virtual environment:

# On Windows
python -m venv venv
.\venv\Scripts\activate

# On macOS/Linux
python3 -m venv venv
source venv/bin/activate

3. Install required packages:

pip install Flask torch torchvision nltk

4. Install NLTK data:

python
>>> import nltk
>>> nltk.download('punkt')

Configuration

1. The chatbot comes pre-configured with CUTM-specific intents in intents.json:

{
    "intents": [
        {
            "tag": "greeting",
            "patterns": ["Hi", "Hello", "Hey"],
            "responses": ["Hello! Welcome to CUTM Chatbot!", "Hi there! How can I help you with CUTM related queries?", "Hey! Ask me anything about Centurion University!"]
        },
        {
            "tag": "courses",
            "patterns": ["What courses are offered?", "Tell me about programs", "Available degrees"],
            "responses": ["CUTM offers various undergraduate and postgraduate programs in Engineering, Management, Pharmacy, and more. Would you like specific information about any program?"]
        }

    ]
}

2. Train the model:

python train.py

3. Verify the training by testing in console:

python chat.py

Running the Application

Option 1: Integrated Flask App

python app.py

Visit http://localhost:5000 in your browser.

Option 2: API-Only Mode

python app.py --api-only

The API will be available at http://localhost:5000/predict.

API Documentation

Prediction Endpoint

• URL: /predict
• Method: POST
• Content-Type: application/json
• Request Body:

{
    "message": "Tell me about CUTM"
}

• Response:

{
    "answer": "Centurion University of Technology and Management is a premier educational institution focused on skill-integrated education."
}

Customization

Adding New Intents

1. Add new patterns and responses to intents.json
2. Retrain the model using train.py
3. Restart the Flask application

Modifying the Neural Network

Edit model.py to customize the neural network architecture:

• Adjust layer sizes
• Modify activation functions
• Add or remove layers

Contributing

1. Fork the repository
2. Create a new branch (git checkout -b feature/improvement)
3. Make changes
4. Commit (git commit -am 'Add new feature')
5. Push (git push origin feature/improvement)
6. Create a Pull Request

Troubleshooting

Common issues and solutions:

1. Model not loading:

   • Ensure data.pth exists
   • Verify PyTorch version compatibility

2. NLTK errors:

   • Run nltk.download('punkt') in Python console
   • Check NLTK installation

3. Flask application not starting:

   • Verify virtual environment is activated
   • Check port 5000 is available

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Centurion University of Technology and Management
• Faculty mentors and guides
• Original PyTorch chatbot tutorial
• Flask documentation
• NLTK community
• PyTorch team

Contact

For questions and support about CUTM CHATBOT, please open an issue in the GitHub repository or contact the development team at Centurion University of Technology and Management.