A Graph Embedding-based Identity Inference Attack on Blockchain Systems
- Title
- A Graph Embedding-based Identity Inference Attack on Blockchain Systems
- Author
- 조성현
- Keywords
- Bioengineering; Communication, Networking and Broadcast Technologies; Components, Circuits, Devices and Systems; Computing and Processing; Engineered Materials, Dielectrics and Plasmas; Fields, Waves and Electromagnetics; Photonics and Electrooptics; Privacy; Data privacy; Machine learning algorithms; Medical services; Feature extraction; Inference algorithms; Blockchains; Blockchain; Smart-Healthcare; Distributed system; Identity inference; Artificial intelligence; Graph classification
- Issue Date
- 2022-02
- Publisher
- IEEE
- Citation
- 2022 International Conference on Electronics, Information, and Communication (ICEIC) Electronics, Information, and Communication (ICEIC), 2022 International Conference on. :1-3 Feb, 2022
- Abstract
- This paper discusses an attack scenario that infers
identity in a blockchain-based smart healthcare system. There
is a privacy vulnerability because blockchain data is open to all
participants. Invasion of privacy is a significant problem in a
blockchain-based smart healthcare system that utilizes personal
information. We analyze the blockchain graph to infer a user’s
identity. Then, we construct account-transaction graphs using
transactions on the blockchain. A graph embedding algorithm
generates features of account-transaction graphs. The generated
features are used to identify blockchain participants using machine learning algorithms. We evaluate our attack scenario by
applying embedding algorithms in Ethereum. The results showed
an inference performance of up to 0.94 in the f1-score.
- URI
- https://ieeexplore.ieee.org/document/9748281?arnumber=9748281&SID=EBSCO:edseeehttps://repository.hanyang.ac.kr/handle/20.500.11754/171055
- ISBN
- 978-1-6654-0934-6
- ISSN
- 2767-7699
- DOI
- 10.1109/ICEIC54506.2022.9748281
- Appears in Collections:
- ETC[S] > 연구정보
- Files in This Item:
There are no files associated with this item.
- Export
- RIS (EndNote)
- XLS (Excel)
- XML