SMS Spam Detector

Overview

This project is a Streamlit application that detects whether an SMS message is spam or non-spam using a pre-trained deep learning model. The model was trained on SMS data and uses natural language processing techniques for prediction. The application provides an intuitive interface for users to input messages and view predictions.

Features

• Interactive Interface: Enter SMS messages to classify them as spam or non-spam.
• Real-time Predictions: Displays the predicted label along with the confidence score.
• Visualization: Pie chart showing the proportion of spam vs non-spam predictions.
• Prediction History: Maintains a log of all predictions made during the session.
• Custom Styling: User-friendly interface with CSS enhancements.

Demo

You can try the application live:

• Docker Deployment on Hugging Face Spaces: SMS Spam Detector - Docker
• Streamlit Cloud Deployment: SMS Spam Detector - Streamlit

Installation

Prerequisites

• Python 3.8 or later
• pip
• A GPU-enabled machine (optional but recommended for TensorFlow)

Steps

1. Clone the repository:

   git clone https://github.com/<your-username>/sms-spam-detector.git
   cd sms-spam-detector

2. Create a virtual environment:

   python -m venv env
   source env/bin/activate  # On Windows: .\env\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

4. Download the pre-trained model and tokenizer:

   • Place the model_wordembed.keras file in the models/ directory.
   • Place the tokenizer_word_index.npy file in the same directory.

5. Run the Streamlit app:

   streamlit run app.py

6. Access the app in your browser at http://localhost:8501.

Project Structure

├── app.py                # Main Streamlit application
├── models/               # Directory for the model and tokenizer
│   ├── model_wordembed.keras
│   ├── tokenizer_word_index.npy
├── requirements.txt      # Python dependencies
└── README.md             # Project documentation

Example Usage

1. Start the application by running streamlit run app.py.
2. Enter a message in the text box.
3. Click the "Predict" button to view the classification result.
4. Check the visualization for the proportion of predictions.

Technologies Used

• Framework: Streamlit
• Machine Learning: TensorFlow, Keras
• Visualization: Plotly
• NLP: Tokenizer, Embedding layers

Future Enhancements

• Add support for additional languages.
• Include training scripts for fine-tuning the model.
• Enhance visualizations with detailed analytics.
• Allow users to choose between multiple pre-trained models within the application.

Author

Christophe Noret

License

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

Dependencies Licenses

• Streamlit: Licensed under the Apache 2.0 License. For details, see Streamlit's GitHub repository.
• TensorFlow: Licensed under the Apache 2.0 License. For details, see TensorFlow's license.
• Plotly: Licensed under the MIT License. For details, see Plotly's GitHub repository.