Phishing URL Detection using Recurrent Weighted Average
Data Privacy is a serious issue with the advent of Technology and prevalence of Social Networks in today's era. This project aims at tackling one avenue how your private data can be compromised - Phishing websites.
The final_dataset.csv contains ~50,000 URLs with 'Benign' and 'Phishing' classes with equal proportion. The Recurrent Weighted Average architecture is used to train a Deep Neural Network to classify these URLs to their respective classes.
The trained model has achieved an accuracy of 98.6% with the current dataset.
Dependencies
- Python 3.0 or higher
- TensorFlow
- Tkinter
Training the Model
-
First clone the repository and install the above dependencies.
-
To start training the model enter the following command in the terminal:
- After training in complete, you can test out any URL by running: