Bidirectional fusion heterogeneous graph networks for semi-supervised Bitcoin transaction anomaly detection in dynamic transaction graphs
Bo Xiao, Wei Yin · 2026
Read Bidirectional fusion heterogeneous graph networks for semi-supervised Bitcoin transaction anomaly detection in dynamic transaction graphs by Bo Xiao, Wei Yin (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0351051.txt.
Description
Bo Xiao and Wei Yin propose a network model to catch fraud in Bitcoin transaction graphs, where accounts and transfers form a shifting, mixed type web and labelled fraud cases are rare. Their model aligns features across node types, fuses time patterns in both directions, and invents extra near real fraud examples to balance training and sharpen the fraud boundary.
On a public labelled Bitcoin data set, scores beat prior methods for both transfer and address fraud flags. Gains came from the invented examples and the two way time fusion most. Exchanges and investigators could triage suspicious flows faster, though live deployment must still prove itself against evolving laundering tricks.
Catalog record
| CREATOR | Bo Xiao, Wei Yin |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 60948 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |