Skip to content

Latest commit

 

History

History
16 lines (12 loc) · 855 Bytes

README.md

File metadata and controls

16 lines (12 loc) · 855 Bytes
Raw

This respository implements three models described in Anti-Money Laundering in Bitcoin: Experimenting with Graph Convolutional Networks for Financial Forensics. Models are applied to Elliptic Bitcoin dataset.

Models

  1. GCN
  2. GCN with skip connection
  3. EvolveGCN

Requirements

  • tensorflow ==2.4.1
  • numpy <= 1.19.5

See articles:

  1. Weber, Mark, et al. "Anti-money laundering in bitcoin: Experimenting with graph convolutional networks for financial forensics." arXiv preprint arXiv:1908.02591 (2019).
  2. Pareja, Aldo, et al. "EvolveGCN: Evolving graph convolutional networks for dynamic graphs." Proceedings of the AAAI Conference on Artificial Intelligence.