基于时序交易图注意力神经网络的以太坊恶意账户检测

doi:10.3969/j.issn.1671-1122.2022.10.010

信息网络安全 ›› 2022, Vol. 22 ›› Issue (10): 69-75.doi: 10.3969/j.issn.1671-1122.2022.10.010

基于时序交易图注意力神经网络的以太坊恶意账户检测

石拓¹, 梁飞²(), 尚钢川², 田洋俊³

1.北京警察学院公安管理系，北京 102202
2.北京市公安局海淀分局警务支援大队，北京 100089
3.金寨县公安局，六安 237351

收稿日期:2022-07-09 出版日期:2022-10-10 发布日期:2022-11-15
通讯作者: 梁飞 E-mail:475662476@qq.com
作者简介:石拓（1988—），女，北京，副教授，博士，主要研究方向为数据挖掘和人工智能|梁飞（1989—），男，北京，工程师，本科，主要研究方向为人工智能和区块链|尚钢川（1990—），男，北京，工程师，本科，主要研究方向为网络安全和人工智能|田洋俊（1988—），男，安徽，本科，主要研究方向为网络安全

Detection of Malicious Ethereum Account Based on Time Series Transaction and Graph Attention Neural Network

SHI Tuo¹, LIANG Fei²(), SHANG Gangchuan², TIAN Yangjun³

1. Department of Public Security Management, Beijing Police College, Beijing 102202, China
2. Haidian Branch Police Support Brigade of Beijing Public Security Bureau, Beijing 100089, China
3. Public Security Bureau of Jinzhai County, Lu’an 237351, China

Received:2022-07-09 Online:2022-10-10 Published:2022-11-15
Contact: LIANG Fei E-mail:475662476@qq.com

摘要/Abstract

摘要：

随着区块链的迅速发展，利用以太坊从事传销、诈骗、洗钱等犯罪行为逐年增加，因此对于以太坊账户的检测成为了破解新型犯罪的一种有效方法。文章提出将交易时间信息融入到以太坊地址账户特征的模型，从而检测以太坊账户是否为恶意账户。模型对传统的注意力网络进行改进，通过融合时序交易时间图注意力的神经网络实现了地址账户特征的最终表达。实验结果表明，该模型优于传统的机器学习分类算法和图神经网络分类算法。

关键词: 图注意力机制, 时间核函数, 以太坊地址

Abstract:

With the rapid development of blockchain, using ethereum to engage in pyramid selling, fraud, and money laundering crimes has increased year by year. Therefore, the detection of ethereum accounts has become an effective method to crack new types of crimes. The information was integrated into the characteristics of the ethereum address and account as a model to detect whether the account was a malicious one. The model in this paper improves the the neural network of graph attention mechanism and the time-series transaction information to realize the final expression of the address account characteristics. It is verified by experiments that the purposed model is superior to the graph neural network classification algorithm established by the traditional classification method.

Key words: graph attention mechanism, time kernel function, ethereum accounts

中图分类号:

TP309

石拓, 梁飞, 尚钢川, 田洋俊. 基于时序交易图注意力神经网络的以太坊恶意账户检测[J]. 信息网络安全, 2022, 22(10): 69-75.

SHI Tuo, LIANG Fei, SHANG Gangchuan, TIAN Yangjun. Detection of Malicious Ethereum Account Based on Time Series Transaction and Graph Attention Neural Network[J]. Netinfo Security, 2022, 22(10): 69-75.

图/表 3

参考文献 15

[1]	MA Chunguang, AN Jing, BI Wei, et al. Smart Contract in Blockchain[J]. Netinfo Security, 2018, 18(11): 8-17.
	马春光, 安婧, 毕伟, 等. 区块链中的智能合约[J]. 信息网络安全, 2018, 18(11): 8-17.
[2]	BIAN Lingyu, ZHANG Linlin, ZHAO Kai, et al. Ethereum Malicious Account Detection Method Based on LightGBM[J]. Netinfo Security, 2020, 20(4): 73-84.
	边玲玉, 张琳琳, 赵楷, 等. 基于LightGBM的以太坊恶意账户检测方法[J]. 信息网络安全, 2020, 20(4):73-84.
[3]	CHEN Weili, ZHENG Zibin, CUI Jiahui, et al. Detecting Ponzi Schemes on Ethereum: Towards Healthier Blockchain Technology[C]// Springer. Proceedings of the 2018 World Wide Web Conference. Heidelberg: Springer, 2018: 1409-1418.
[4]	TAO Feng, WANG Jinsong, LIU Duobin, et al. Bitcoin User Address Classification Method Based on Improved Random Forest[J]. Journal of Tianjin University of Technology, 2022, 38(1): 53-58.
	陶峰, 王劲松, 吕垛斌, 等. 基于改进随机森林的比特币用户地址分类方法[J]. 天津理工大学学报, 2022, 38(1): 53-58.
[5]	LIU Xiao, TANG Zaiyang, LI Peng, et al. A Graph Learning Based Approach for Identity Inference in DApp Platform Blockchain[J]. IEEE Transactions on Emerging Topics in Computing, 2022, 10(1): 438-449.
[6]	KANEZASHI H, SUZUMURAT, LIU Xin, et al. Ethereum Fraud Detection with Heterogeneous Graph Neural Networks[EB/OL]. (2022-07-04)[2022-07-08]. https://arxiv.org/abs/2203.12363.
[7]	VELICKOVIC P, CUCURULL G, CASANOVA A, et al. Graph Attention Networks[C]// Springer. International Conference on Learning Representations. Heidelberg: Springer, 2018: 1050-1062.
[8]	XU D, RUAN C, KORPROGLU E, et al. Inductive Representation Learning on Temporal Graphs[EB/OL]. (2020-10-19)[2022-07-08]. https://arxiv.org/abs/2002.07962.
[9]	ZHOU Jun, ZHANG Zhiqiang, CAO Yuetian, et al. A Geography and Time Aware Representation Learning Framework[J]. CAAI Transactions on Intelligent Systems, 2021, 16(5): 909-917.
	周俊, 张志强, 曹月恬, 等. 地理位置和时间感知的表示学习框架[J]. 智能系统学报, 2021, 16(5): 909-917
[10]	CHEN Liang, PENG Jiaying, LIU Yang, et al. Phishing Scams Detection in Ethereum Transaction Network[J]. ACM Transactions on Internet Technology, 2021, 21(1): 1-16.
[11]	ZHOU Jiajun, HU Chenkai, CHI Jianlei, et al. Behavior-Aware Account De-Anonymization on Ethereum Interaction Graph[EB/OL]. [2022-07-08]. https://arxiv.org/abs/2203.09360.
[12]	LI Panpan, XIE Yunyi, XU Xinyao, et al. Phishing Fraud Detection on Ethereum Using Graph Neural Network[EB/OL]. (2022-04-18)[2022-07-08]. https://arxiv.org/abs/2204.08194
[13]	KINGMA P, BA J. Adam: A Method for Stochastic Optimization[EB/OL]. (2017-01-30)[2022-07-08]. https://arxiv.org/abs/1412.6980.
[14]	LIANG Fei. Intelligent Detection Method of Malicious URL Based on Multi Group Attention Mechanism[J]. Netinfo Security, 2020, 20(1): 18-22.
	梁飞. 基于多分组注意力机制的恶意URL智能检测方法[J]. 信息网络安全, 2020, 20(1):18-22.
[15]	WU Jiajing, LIU Jieli, LIN Dan, et al. Blockchain Transaction Networks:A Survey[J]. Acta Scientiarum Naturalium Universitatis Sunyatseni, 2021, 60(5): 1-12.
	吴嘉婧, 刘洁利, 林丹, 等. 区块链交易网络研究综述[J]. 中山大学学报(自然科学版), 2021, 60(5):1-12.

检测算法	Precision	Recall	F1
LightGBM	82.63%	62.58%	71.22%
Graph-Attention	86.25%	68.32%	76.25%
Time-Graph-Att	88.76%	78.56%	83.35%

基于时序交易图注意力神经网络的以太坊恶意账户检测

Detection of Malicious Ethereum Account Based on Time Series Transaction and Graph Attention Neural Network

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 3

参考文献 15

相关文章 1

编辑推荐

Metrics

本文评价