🤖 AI-Based Customer Satisfaction Tracker

This project is a smart tool that evaluates and tracks customer satisfaction during conversations using LLMs from OpenRouter.ai. It provides:

• A web app interface built with Streamlit
• A backend tracking engine using Python + LLM
• Real-time score updates for user messages

---

🚀 Features

• Tracks customer satisfaction (scale: 1 to 5)
• Ignores greetings or empty messages
• Uses LLM to interpret tone/meaning
• Interactive UI for chatting and feedback
• Built-in satisfaction scoring logic

---

📁 Project Structure

.
├── app.py                    # Main Streamlit app interface
├── satisfaction_tracker.py   # Satisfaction tracker class
├── server.py                 # (Optional) server logic
├── requirements.txt          # Python dependencies
└── README.md                 # Documentation

---

🔧 Setup Instructions

1. 🔁 Clone the Repository

git clone https://github.com/Hackathon4-Team20/ai-model.git
cd ai-model

Alternatively: Download the ZIP file and extract it.

2. 🐍 Create a Virtual Environment (Recommended)

python -m venv venv

# Activate virtual environment:
# On Windows:
venv\Scripts\activate
# On macOS/Linux:
source venv/bin/activate

3. 📦 Install Dependencies

pip install -r requirements.txt

4. 🔐 API Key Setup

You'll need an API key from OpenRouter:

1. When the app starts, it will prompt you for the API key
2. Paste your key in the input field (never commit this to the repository)

5. ▶️ Running the Application

Launch the server with:

python server.py

---

🧠 How It Works

1. You chat via the web app interface
2. Your messages are analyzed using OpenRouter's LLM
3. A satisfaction score (1–5) is calculated and displayed in real-time

---

---

⚠️ Important Notes

• Never commit your API key to the repository
• For production use, consider:
  • Using environment variables (.env file)
  • Implementing proper security measures
  • Rate limiting your API calls

---

💡 Additional Options (Available on Request)

I can help you add:

• .env setup examples
• Docker support
• Hosting instructions (Streamlit Cloud, etc.)
• Collaboration guidelines

Let me know if you'd like to enhance this setup for deployment or team collaboration!