Graph Convolutional Network (GCN) model for ex-ante detection of cybercrimes on the Ethereum blockchain


The code implements a Graph Convolutional Network (GCN) model to classify cybercriminal and non-cybercriminal addresses on the Ethereum blockchain, using ATGraph technology. Our approach uses several on-chain input features for classification. It builds on Hornuf, Momtaz, Nam, and Ye (2024). and is designed to perform well in samples of limited size.

Click here for the Python implementation.