基于交易不可信度的比特币非法交易检测

doi:10.19678/j.issn.1000-3428.0062816

摘要/Abstract

摘要： 比特币非法交易检测根据交易特征构建检测模型鉴别非法交易，在反金融犯罪领域得到广泛应用。现有的比特币非法交易检测方法假定交易的属性包含交易是否非法的信息，并从交易属性中挖掘能够代表交易非法性的特征，难以准确判断非法交易，导致检测精度和召回率降低。设计基于交易不可信度的比特币非法交易检测方法。根据非法交易之间具有关联的特性，定义交易不可信度。通过构建交易不可信度度量模型，将量化结果作为直接反映交易非法性的特征融入到已有的分类模型中，提高模型的检测性能。在此基础上，采用迭代训练集的方式扩增非法交易样本，解决非法交易样本不足以及标注困难的问题。在Elliptic数据集上的实验结果表明，与本地特征和聚合特征相比，加入不可信度特征的逻辑回归、随机森林、多层感知机和图卷积网络分类模型的F1值平均提高8.5%。

关键词: 比特币, 区块链, 非法交易, 机器学习, 图卷积网络

Abstract: Detection of illicit Bitcoin transaction builds a detection model according to transaction features to identify illicit transactions.It has been widely used in the field of anti-financial crime.Existing method for the detection of illicit Bitcoin transaction proposes that the transaction attributes contain information about whether the transaction is illicit and mine the features that can represent the illicitness of the transaction from the transaction attributes.It is difficult to accurately judge illicit transactions, resulting in the reduction of detection accuracy and recall.This paper proposes a method for the detection of illicit Bitcoin transaction based on transaction unreliability.According to the features of the correlation between illicit transactions, transaction unreliability is proposed as a measure of the degree of transaction unreliability.By constructing the measurement model of transaction unreliability and integrating the quantitative results into the existing classification model as features that directly reflect the illicit nature of transactions, the detection ability of the model improved.On this basis, the illicit transaction samples are expanded using an iterative training set to solve the problems of insufficient illicit transaction samples and difficult labeling.The experiments on Elliptic data sets show that compared with local features and aggregation features, the F1 value of Logistic Regression(LR), Random Forest(RF), MultiLayer Perceptron(MLP), and Graph Convolutional Network(GCN) classification models with unreliability features increased by 8.5%, on average.

Key words: Bitcoin, blockchain, illicit transaction, machine learning, Graph Convolutional Network(GCN)

中图分类号:

TP391

俞莎莎, 牛保宁. 基于交易不可信度的比特币非法交易检测[J]. 计算机工程, 2022, 48(8): 166-172.

YU Shasha, NIU Baoning. Detection of Illicit Bitcoin Transaction Based on Transaction Unreliablity[J]. Computer Engineering, 2022, 48(8): 166-172.

http://www.ecice06.com/CN/Y2022/V48/I8/166

图/表 6

20220825092218

20220825092222

20220825092231

20220825092235

20220825092238

20220825092242

参考文献

[1] SQUAREPANTS S.Bitcoin:a peer-to-peer electronic cash system[EB/OL].[2021-08-25].https://courses.cs.washington.edu/courses/csep552/18wi/papers/nakamoto-bitcoin.pdf.
[2] 张亮, 刘百祥, 张如意, 等.区块链技术综述[J].计算机工程, 2019, 45(5):1-12. ZHANG L, LIU B X, ZHANG R Y, et al.Overview of blockchain technology[J].Computer Engineering, 2019, 45(5):1-12.(in Chinese)
[3] ATAMER I.Bitcoin scams &cryptocurrency fraud[EB/OL].[2021-08-25].https://arxiv.org/abs/1758.02591.
[4] WEBER M, DOMENICONI G, CHEN J, et al.Anti-money laundering in Bitcoin:experimenting with graph convolutional networks for financial forensics[EB/OL].[2021-08-25].https://arxiv.org/abs/1908.02591.
[5] CRAWFORD J, GUAN Y.Knowing your Bitcoin customer:money laundering in the Bitcoin economy[C]//Proceedings of the 13th International Conference on Systematic Approaches to Digital Forensic Engineering.Washington D.C., USA:IEEE Press, 2020:1-10.
[6] SICIGNANO G J.Money laundering using cryptocurrency:the case of Bitcoin![EB/OL].[2021-08-25].https://arxiv.org/abs/1908.02911.
[7] HOUT M V, BINGHAM T.Responsible vendors, intelligent consumers:silk road, the online revolution in drug trading[J].Int J Drug Policy, 2014, 25(2):183-189.
[8] GODSIFF P.Bitcoin:bubble or blockchain[M].Berlin, Germany:Springer International Publishing, 2015.
[9] HARLEV M A, YIN H S, LANGENHELDT K C, et al.Breaking bad:de-anonymising entity types on the Bitcoin blockchain using supervised machine learning[C]//Proceedings of the 51st Hawaii International Conference on System Sciences.Washington D.C., USA:IEEE Press, 2018:1-10.
[10] MENARD S.Logistic regression[J].American Statistician, 2004, 58(4):364.
[11] BREIMAN L.Random forests, machine learning 45[J].Journal of Clinical Microbiology, 2001, 2:199-228.
[12] AHR M, BIEHL M, SCHLÖSSER E.Multilayer neural networks[M].Berlin, Germany:Springer, 2004.
[13] KIPF T N, WELLING M.Semi-supervised classification with graph convolutional networks[EB/OL].[2021-08-25].https://arxiv.org/pdf/1609.02907.pdf.
[14] PAREJA A, DOMENICONI G, CHEN J, et al.EvolveGCN:evolving graph convolutional networks for dynamic graphs[EB/OL].[2021-08-25].https://arxiv.org/pdf/1902.10191.pdf.
[15] PATIL V R, NIKAN A P, PAWAR J S, et al.Bitcoin fraud detection using data mining approach[J].Journal of Information Technology and Sciences, 2018, 4(2):1-10.
[16] ANDROULAKI E, KARAME O G, ROESCHLIN M, et al.Evaluating user privacy in Bitcoin[C]//Proceedings of International Conference on Financial Cryptography and Data Security.Berlin, Germany:Springer, 2012:34-51.
[17] MEIKLEJOHN S, POMAROLE M, JORDAN G, et al.A fistful of Bitcoins:characterizing payments among men with no names[J].Communications of the ACM, 2016, 59(4):86-93.
[18] SUN YIN H H, LANGENHELDT K, HARLEV M, et al.Regulating cryptocurrencies:a supervised machine learning approach to de-anonymizing the Bitcoin blockchain[J].Journal of Management Information Systems, 2019, 36(1):37-73.
[19] LIN Y J, WU P W, HSU C H, et al.An evaluation of Bitcoin address classification based on transaction history summarization[EB/OL].[2021-08-25].https://arxiv.org/pdf/1903.07994.pdf.
[20] JOURDAN M, BLANDIN S, WYNTER L, et al.Characterizing entities in the Bitcoin blockchain[C]//Proceedings of IEEE International Conference on Data Mining Workshops.Washington D.C., USA:IEEE Press, 2018:1-10.
[21] JUHÁSZ P L, STÉGER J, KONDOR D, et al.A Bayesian approach to identify Bitcoin users[EB/OL].[2021-08-25].https://arxiv.org/abs/1612.06747.
[22] WU J, LIU J, CHEN W, et al.Detecting mixing services via mining Bitcoin transaction network with hybrid motifs[EB/OL].[2021-08-25].https://arxiv.org/abs/2001.05233.
[23] ANZAI Y.Pattern recognition and machine learning[M].Amsterdam, Netherlands:Elsevier, 1992.
[24] 张立军, 袁能文.线性综合评价模型中指标标准化方法的比较与选择[J].统计与信息论坛, 2010, 25(8):10-15. ZHANG L J, YUAN N W.Comparison and selection of index standardization method in linear comprehensive evaluation model[J].Statistics &Information Forum, 2010, 25(8):10-15.(in Chinese)

选择文件类型/文献管理软件名称

选择包含的内容