Scalable Multi-chain Graph-based User Profiling Framework for EVM-Compatible Blockchains

The ultimate level of privacy is a double-edged sword. On the one hand, it guards honest users’ private information; while benefiting criminals to perform fraudulent activities anonymously. Blockchain as a decentralized environment is not prone to this problem. Decentralization enables bad actors to conduct wash trading, money laundering, tax evasion, scam, and fraud, to mention a few. Previous studies only focused on detecting a limited set of fraudulent activities. However, the primary purpose of this work is to make a framework that facilitates building different types of forensic tools for all EVM-compatible blockchains. The project will be done in several phases, including labeling and classifying addresses using state-of-the-art Machine Learning (ML) and graph-based solutions.

Faculty Supervisor:

Victoria Lemieux;Chen Feng

Student:

Partner:

Covalent

Discipline:

Computer science

Sector:

Information and cultural industries

University:

The University of British Columbia

Program: