A forensic analysis of a graph dataset reveals clusters associated with illicit activity and money laundering on the Bitcoin blockchain. The analysis uncovers previously unknown wallets belonging to a Russian darknet market and criminal proceeds sent to a crypto exchange. Machine learning at the subgraph level is effective in predicting whether crypto transactions constitute proceeds of crime. The study experiments with different subgraph classification methods, tracing back the source of funds associated with suspicious subgraphs to various entities. Further examination identifies known cryptocurrency laundering patterns. Future research will focus on increasing accuracy and extending the work to other blockchains.
12 Impressions