ScamGuard is a Flutter-based application that integrates a machine learning model to detect scams based on user-provided input. This project was developed as part of a university course on machine learning and serves as an example of deploying machine learning models in real-world applications.
-
Machine Learning Integration:
- The app uses a Random Forest Tree model for prediction, as it provides high accuracy.
- The trained models are serialized using Python's joblib library to
.pklfiles for easy backend integration.
-
Backend and Frontend Communication:
- A Flask backend serves as the bridge between the machine learning model and the Flutter frontend.
- The backend returns predictions in JSON format, which are then displayed in the app.
-
User Authentication (Optional):
- Users can register and log in through the app.
- User data is securely stored in Firebase Cloud Database.
- Flutter: Cross-platform framework for the mobile app.
- Python: For model integration and backend logic.
- Flask: Lightweight Python web framework for serving the API.
- Firebase: For user authentication and data storage.
- Random Forest Tree: A supervised learning algorithm for predictions.
- joblib (Python): Used to serialize and deserialize the trained models.
-
Train the Model:
- The model is trained using a dataset in Python.
- Random Forest Tree is used for predictions, and the final model is exported as a
.pklfile using joblib.
-
Setup the Backend:
- A Flask application (
app.py) loads the.pklmodel and serves an API endpoint for predictions. - The backend processes user input and returns predictions in JSON format.
- A Flask application (
-
Connect to the Frontend:
- The Flutter app fetches data from the Flask backend via HTTP requests.
- Predictions are displayed in the app in a user-friendly format.
-
User Authentication:
- Optional user registration and login features are implemented using Firebase.
- Flutter Development Environment: Get started with Flutter.
- Python Environment: Install Python 3.7+ with Flask and joblib libraries.
- Clone the Repository:
git clone https://github.com/your-username/scamguard.git cd scamguard **pip install flask joblib** **python app.py** **flutter pub get** **flutter run**
This version includes:
- Structured Layout: Clear sections for features, technologies, workflow, and setup instructions.
- Professional Tone: Improves readability and conveys a professional approach.
- Optional Sections: Allows flexibility to include screenshots or other project details.
- Getting Started Instructions: Provides step-by-step guidance for new developers.