🧠 Spam Detector - AI-Powered Message Classification

Detect spam messages with a simple and elegant web interface powered by machine learning.

🚀 Features

• 🔍 Real-time spam detection using a trained LSTM + GRU neural network
• 🧠 Built with TensorFlow and Keras
• 🌐 Interactive frontend with Tailwind CSS
• 🗂 Flask API for backend prediction
• 🔁 Live result display with modern UI

---

📁 Project Structure

/spam-detector
├── app.py                   # (Optional) Model training script
├── predict_spam.py         # Flask API for serving predictions
├── tokenizer.pickle        # Saved tokenizer used during training
├── spam_detection_model.h5 # Trained spam detection model
├── SpamTrain.csv           # Training data
├── SpamTest.csv            # Test data
├── templates/
│   ├── index.html          # Frontend input page
│   └── result.html         # Result display page
├── static/                 # Optional styles/scripts

---

⚙️ How to Run

1. Install dependencies

   pip install flask tensorflow flask-cors

2. Start the Flask server

   python predict_spam.py

3. Launch the frontend

   python -m http.server 8000

4. Visit in browser

   http://localhost:8000/index.html
   

---

💡 How It Works

• User types a message into the frontend
• The input is sent to the Flask API via a POST request
• The server tokenizes and classifies the message as "Spam" or "Not Spam"
• The result is shown on a separate result page

---

🛠 Tech Stack

• Frontend: HTML, Tailwind CSS, JavaScript
• Backend: Python, Flask, TensorFlow
• Model: LSTM + GRU hybrid for spam classification
• Deployment Ready: Can be served locally or hosted via Flask + frontend hosting (Netlify, Vercel, etc.)

---