SigTran: Signature Vectors for Detecting Illicit Activities in Blockchain Transaction Networks

This repository provides the implementation of SigTran as explained in the paper:

"SigTran: Signature Vectors for Detecting Illicit Activities in Blockchain Transaction Networks." Farimah Poursafaei, Reihaneh Rabbany, and Zeljko Zilic. The 25th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD-2021), 2021.

The paper can be found here.

Overview

This repository contains the implementation code of SigTran which is an efficient graph-based method for detecting illicit nodes on blockchain networks. Please refer to our paper for more details.

Data

We conducted our experiments on two public datasets including transaction histories of Bitcoin and Ethereum blockchain.

Acknowledgement

We would like to thank the authors of RiWalk and node2vec for providing open access to the implementation of their methods.

Citation

If this work is useful for your research, please cite us:

@inproceedings{sigtran_pakdd_2021,
  title={SigTran: Signature Vectors for Detecting Illicit Activities in Blockchain Transaction Networks},
  author={Poursafaei, Farimah and Rabbany, Reihaneh and Zilic, Zeljko},
  booktitle={Pacific-Asia Conference on Knowledge Discovery and Data Mining},
  pages={27--39},
  year={2021},
  organization={Springer}
}