Skip to content
AI: Deep Learning for Phishing URL Detection
Python
Branch: master
Clone or download
Latest commit f2d6910 Dec 7, 2018
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
LICENSE Create LICENSE Dec 3, 2018
README.md Create README.md Dec 7, 2018
flaskrestapi.py Update flaskrestapi.py Dec 4, 2018
label_data.py Initial commit. Dec 2, 2018
phishing_database.csv Initial commit. Dec 2, 2018
request.py Update request.py Dec 4, 2018
requirements.txt Initial commit. Dec 2, 2018
train.py Initial commit. Dec 2, 2018
whitelist.txt Initial commit. Dec 2, 2018

README.md

AI: Deep Learning for Phishing URL Detection

Model Performance

ROC/AUC Curve Confusion Matrix F1 Score

Requirements

This code was created with Python 3.6.7. Other versions of Python 3 might also work. You can have multiple Python versions (2.x and 3.x) installed on the same system without problems.

Make sure to install all requirements:

$ pip install -r requirements.txt

NOTICE : Because of Github size limits, please download the model from here: https://www.zpettry.com/bi-lstmchar256256128.h5

Quick start

Ensure the model has been downloaded from the above link.

Open a separate tab or window and run:

$ python3 flaskrestapi.py

Now go back to the original tab or window and run:

$ python3 request.py -u https://www.google.com/about

Output:

$ [{'malicious percentage': 2.552182786166668, 'result': 'URL is probably NOT malicious.', 'url': 'https://www.google.com/about'}]

Web site and documentation

Blog and additional information about this project is available at the web site:

https://www.zpettry.com/

License

This code is licensed under the terms of the MIT License (see the file LICENSE).

You can’t perform that action at this time.