Skip navigation Jump to main navigation Jump to main navigation

Fall Update

At SPS this fall, all courses, other than pre-established online courses, will be offered face-to-face in our New York City classrooms. Some of these face-to-face courses will be offered in the HyFlex format to ensure that all of our students can make progress toward their degree requirements, if faced with delays due to student visas or vaccination effectiveness wait times.
Close alert

Protecting Against Ransomware: Professor Siddhartha Dalal’s Research on Identifying Ransomware Actors in the Bitcoin Network

How organizations can protect themselves from ransomware attacks is a critical issue of the day. 

After an increase in ransomware attacks, the Justice Department recently announced it will give the same level of attention to the threat as it does to terrorism.  "Ransomware is part of an emerging and profitable criminal business that generated more than $400 million in income in 2020,” reports The Wall Street Journal. “FBI Director Christopher Wray said the agency was investigating about 100 different types of ransomware, many of which trace back to hackers in Russia, and compared the current spate of cyberattacks with the challenge posed by the Sept. 11, 2001, terrorist attacks.”

Professor of Professional Practice in Applied Analytics Siddhartha Dalal has co-authored a paper, “Identifying Ransomware Actors in the Bitcoin Network,” along with two Columbia University students, Siddhanth Sabharwal and Zihe Wang. The study offers systematic ways to identify fraudulent actors using graph classification. 

Due to the pseudo-anonymity of the Bitcoin network, users can hide behind their bitcoin addresses that can be generated in unlimited quantity, on the fly, without any formal links between them,” reads the abstract. The paper shows that one can identify common patterns associated with these fraudulent activities and apply them to find other ransomware actors. It creates and applies new AI algorithms for local clustering and supervised graph machine learning for identifying malicious actors. 

The paper showcases how one can extract and create graphs that showcase transaction activities of these miscreant actors. The paper represents significant advances in research on new models that organizations and law enforcement agencies can deploy for identification of ransomware actors and the paths they took to anonymize their identity.  

The paper was accepted for presentation at the 2nd International Conference on Machine Learning, IOT and Blockchain (MLIOB 2021) in Chennai, India, scheduled to take place August 21st - 22nd, 2021.