SharifAmit/Semi-supervised-Phishing-Detection-GAN

[Tensorflow] A Game Theoretic approach using GAN for Phishing URL synthesis and detection

Semi-supervised-Phishing-Detection-GAN

This code is for our paper "Semi-supervised Conditional GAN for Simultaneous Generation and Detection of Phishing URLs: A Game theoretic Perspective" which is under review.

Arxiv Pre-print

https://arxiv.org/abs/2108.01852

Citation

@article{kamran2021semi,
  title={Semi-supervised Conditional GAN for Simultaneous Generation and Detection of Phishing URLs: A Game theoretic Perspective},
  author={Kamran, Sharif Amit and Sengupta, Shamik and Tavakkoli, Alireza},
  journal={arXiv preprint arXiv:2108.01852},
  year={2021}
}

Pre-requisite

• Ubuntu 18.04 / Windows 7 or later
• NVIDIA Graphics card

Installation Instruction for Ubuntu

• Download and Install Nvidia Drivers
• Download and Install via Runfile Nvidia Cuda Toolkit 10.0
• Download and Install Nvidia CuDNN 7.6.5 or later
• Install Pip3 and Python3 enviornment

sudo apt-get install pip3 python3-dev

• Install Tensorflow-Gpu version-2.0.0 and Keras version-2.3.1

sudo pip3 install tensorflow-gpu==2.0.0
sudo pip3 install keras==2.3.1

• Install packages from requirements.txt

sudo pip3 install -r requirements.txt

Dataset download link for Phishing URLs

https://www.kaggle.com/taruntiwarihp/phishing-site-urls

NPZ file conversion

• Preprocess all the data to npz format using data_preprocess.py file.

python3 data_preprocess.py --url_length=200 --npz_filename='phishing.npz' --n_samples=50000

• There are different flags to choose from. Not all of them are mandatory.

    '--url_length', type=int, default=200
    '--npz_filename', type=str, default='phishing.npz'
    '--n_sampels',types=int, default=50000,help='number of good and bad samples.'

Training

• Type this in terminal to run the train.py file

python3 train.py --npz_file=phishing.npz --batch_size=64 --epochs=200 --savedir=PhishGan --resume_training=no --latent_dim=50

• There are different flags to choose from. Not all of them are mandatory

   '--epochs', type=int, default=200
   '--batch_size', type=int, default=64
   '--npz_file', type=str, default='phishing.npz', help='path/to/npz/file'
   '--latent_dim', type=int, default=50
   '--savedir', type=str, required=False, help='path/to/save_directory',default='PhishGan'
   '--resume_training', type=str, required=False,  default='no', choices=['yes','no']
   '--weight_name_dis',type=str, help='path/to/discriminator/weight/.h5 file', required=False
   '--weight_name_gen',type=str, help='path/to/generator/weight/.h5 file', required=False