The Shape of Money Laundering: Subgraph Representation Learning on the Blockchain with the Elliptic2 Dataset

By Claudio Bellei, Muhua Xu, Ross Phillips, Tom Robinson, Mark Weber, Tim Kaler, Charles E. Leiserson, Arvind, Jie Chen

Rating

1641
Battle Count: 62

Relevance

6/10
While not directly applicable to trading, the techniques could be adapted for detecting anomalous patterns in financial transaction networks

Implementation Complexity

8/10
Requires handling of large-scale graph data and implementation of complex graph neural network architectures

Reproducibility

4/5
Dataset and code are publicly available, but full reproduction may require significant computational resources

About this paper

Methodology: Subgraph Representation Learning. Problem types: Binary Classification, Subgraph Classification.

The interactive Everscope explorer (charts, battles, favorites) loads below.