Dual-channel Early Warning Framework for Ethereum Ponzi Schemes, Available at link.
Recent versions of the following packages for Python are required:
- toch 1.10.1+cu102
- toch-geometric-temporal 0.50.0
- scikit-learn 1.0.2
- numpy 1.21.5
- pandas 1.3.5
- networkx 2.6.3
PUP_node_dataset : Collect and integrate tagged and transaction-logged Ethereum Ponzi contracts , and randomly acquired non-Ponzi contracts. That is, the Ponzi and non-Ponzi contract datasets
data_new\data_no_aug: The experimental dataset enhanced with THAug and the dataset without enhancement, respectively
baseline：Detection models based on machine learning methods
model: A detection model based on the graph neural network approach, which also defines our model
utils：Toolkit for handling data and defining parameters
main：The main program that executes the runs of our model
@article{jin2022dual,
title={Dual-channel Early Warning Framework for Ethereum Ponzi Schemes},
author={Jin, Jie and Zhou, Jiajun and Jin, Chengxiang and Yu, Shanqing and Zheng, Ziwan and Xuan, Qi},
journal={arXiv preprint arXiv:2206.07895},
year={2022}
}