Illicit Bitcoin Wallets’ Modi Operandi and Sub-Network Discovery using Stochastic Block Modeling

Financial crime in cryptocurrency networks remains difficult to manage as illicit actors camouflage themselves in the noisy open network and rely on complex obfuscation patterns or modi operandi. Tested on a real Bitcoin dataset of 49 snapshots in time, this paper evaluates whether nested, degree-corrected stochastic block modeling can effectively reveal the sub-networks composed of illicit actors beyond simple graph connectivity or numerical attributes and demonstrates the practical applicability of such models for forensic analytics by significantly scoping wide, noisy networks into actionable blocks based on the behavior of wallets.