Machine learning methods to detect money laundering in the Bitcoin blockchain in the presence of label scarcity

TL;DR

This paper presents an active learning approach for detecting Bitcoin money laundering with minimal labeled data, outperforming unsupervised methods and matching fully supervised models using only 5% labels.

Contribution

It introduces an active learning framework that effectively detects illicit Bitcoin transactions with limited labels, addressing label scarcity in money laundering detection.

Findings

Active learning matches fully supervised performance with 5% labels.

Unsupervised anomaly detection methods are inadequate for real Bitcoin data.

Proposed method reduces labeling effort significantly.

Abstract

Every year, criminals launder billions of dollars acquired from serious felonies (e.g., terrorism, drug smuggling, or human trafficking) harming countless people and economies. Cryptocurrencies, in particular, have developed as a haven for money laundering activity. Machine Learning can be used to detect these illicit patterns. However, labels are so scarce that traditional supervised algorithms are inapplicable. Here, we address money laundering detection assuming minimal access to labels. First, we show that existing state-of-the-art solutions using unsupervised anomaly detection methods are inadequate to detect the illicit patterns in a real Bitcoin transaction dataset. Then, we show that our proposed active learning solution is capable of matching the performance of a fully supervised baseline by using just 5\% of the labels. This solution mimics a typical real-life situation in which a limited number of labels can be acquired through manual annotation by experts.