
The proposal refines address clustering with wallet fragments and confidence labeling, drawing on the Bitcoin Fog case where a judge found Chainalysis Reactor “highly reliable” after a Daubert hearing.
Chainalysis proposed an ontology-based methodology to bring more consistent standards to on-chain fund tracing, a process used by law enforcement and investigators to follow digital asset flows across blockchain networks. The framework covers address clustering, wallet segmentation, functional roles, transaction graph structure, and inference confidence, and further separates clusters into finer structures such as wallet fragments while adding a second layer for confidence labeling. Chainalysis said the approach draws on case experience including the Bitcoin Fog proceeding, in which a judge found its Reactor tool “highly reliable” after a Daubert hearing. The company is making the methodology available for wider industry discussion.