Visible to the public CAREER: Cryptocurrency Forensics ToolsConflict Detection Enabled

Project Details

Lead PI

Performance Period

Feb 15, 2019 - Jan 31, 2024

Institution(s)

New York University

Award Number


Cryptocurrencies, such as Bitcoin, are growing in popularity. These cryptocurrencies offer the promise of increased efficiency and decreasing frictions in the financial system, such as international money transfer fees and costs associated with raising investment capital. Unfortunately, they are also misused as a payment mechanism for illicit activities such as extortion, drugs, human trafficking, and cybercrime. These illicit activities have likely diminished the reputation of these cryptocurrencies and facilitated large amounts of harm for entities and individual people. Improved forensic tools that can potentially de-anonymize illicit transactions or capture valuable semantic information about a transaction (for example, what was purchased in a particular transaction), could reduce both of these negative effects. This project will conduct open research that will improve our understanding of how to devise improved cryptocurrency forensic techniques that can be adopted by researchers, companies, and investigators. Much of the research into advanced cryptocurrency forensic techniques is currently being performed by companies and integrated into their closed platforms. This has resulted in a lack of public, generalizable understanding of how to detect and understand the illicit activities occurring within these cryptocurrency ecosystems. The advanced cryptocurrency forensics tools developed in this project will be open and enable improved detection of illicit cryptocurrency transactions. This project will also provide students with expertise in financial technology, security, and machine learning which are all areas of broad national needs.

Achieving the goal of improving the efficacy of cryptocurrency forensics tools requires progress on several key challenges. These include: (1) Designing and implementing improved and open cryptocurrency forensic data collection and archiving systems. (2) Investigating cryptocurrency address labeling techniques. (3) Designing algorithms for improved clustering of addresses and flow tracing. The approach proposed in this project is unique in that it blends improvements to data collection and archiving with advanced machine learning-based algorithms. The project will deepen the understanding of how to effectively perform forensics of a broad range of cryptocurrencies. The ability to effectively forensically analyze cryptocurrencies fundamentally affects our ability to understand and mitigate illicit activities that impact many United States citizens.