This is the PyTorch implementation for the paper BERT4ETH: A Pre-trained Transformer for Ethereum Fraud Detection, accepted by the ACM Web conference (WWW) 2023.

The implementation of fine-tuning on phishing data was finished by Qiustander (https://github.com/Qiustander). Thanks a lot!

If you find this repository useful, please give us a star and cite our paper : ) Thank you!

PyTorch > 1.12.0

• Transaction Dataset:

• • Phishing Account
• • De-anonymization(ENS)
• • De-anonymization(Tornado)
• • Normal Account

• ERC-20 Log Dataset (all in one)

cd BERT4ETH_PyTorch/data; # Labels are already included unzip ...;

cd src; python gen_seq.py --bizdate=bert4eth_exp

python run_pretrain.py --bizdate="bert4eth_exp" \ --ckpt_dir="bert4eth_exp"

python run_embed.py --bizdate="bert4eth_exp" \ --init_checkpoint="bert4eth_exp/xxx.pth"

cd eval python phish_detection_mlp.py --input_dir="../outputs/xxx"

python run_dean_ENS.py --metric=euclidean \ --init_checkpoint=bert4eth_exp/model_104000

 Will update later..

@inproceedings{hu2023bert4eth, title={BERT4ETH: A Pre-trained Transformer for Ethereum Fraud Detection}, author={Hu, Sihao and Zhang, Zhen and Luo, Bingqiao and Lu, Shengliang and He, Bingsheng and Liu, Ling}, booktitle={Proceedings of the ACM Web Conference 2023}, pages={2189--2197}, year={2023} } 

If you have any questions, you can either open an issue or contact me (sihaohu@gatech.edu), and I will reply as soon as I see the issue or email.