以太坊非法交易检测方法综述

李梦; 梁广俊; 印杰; 马卓; 张祎

引用本文：

李梦,梁广俊,印杰,马卓,张祎.以太坊非法交易检测方法综述[J].信息安全学报,已采用 [点击复制]
Li Meng,Liang Guangjun,Yin Jie,Ma Zhuo,Zhang Yi.A Survey of Ethereum Illegal Detection Methods[J].Journal of Cyber Security,Accept [点击复制]

本文已被：浏览 654次下载 0次
以太坊非法交易检测方法综述
李梦¹, 梁广俊¹, 印杰¹, 马卓¹, 张祎²
0 字体:加大+\|默认\|缩小-
(1.江苏警官学院;2.江苏省公安厅)

摘要:

本文针对以太坊非法交易的检测方法展开研究，首先基于以太坊交易数据的特点对非法交易的通用检测方法进行了总结，梳理发现主流算法借助监督算法进行用户地址分类，借助无监督聚类算法发现潜在非法用户。然后针对网络诈骗、庞氏骗局和蜜罐合约这三种特定的诈骗类型，分别总结其在以太坊平台上体现出的“新特点”，并总结了利用机器学习和图嵌入学习等算法构建检测模型的相关研究思路、建立过程及实验性能评估。最后，对以太坊非法交易检测给出了未来的研究方向。

关键词: 以太坊非法检测机器学习庞氏骗局蜜罐合约网络钓鱼诈骗

DOI：

投稿时间：2022-11-16修订日期：2023-03-28

基金项目:国家自然科学基金青年基金（62202209）,南京邮电大学射频集成与微组装技术国家地方联合工程实验室开放课题（KFJJ20200201），江苏省教育厅科研项目（2021SJA0497），江苏警官学院科研项目（2020SJYZR02）

A Survey of Ethereum Illegal Detection Methods

Li Meng¹, Liang Guangjun¹, Yin Jie¹, Ma Zhuo¹, Zhang Yi²

(1.Jiangsu Police Institute;2.Department of Jiangsu Provincial Public Security)

Abstract:

This paper studies the detection method of illegal transactions in Ethereum. First, based on the characteristics of Ethereum transaction data, the general detection methods for illegal transactions are summarized. We found that mainstream algorithms use supervised algorithms to classify user addresses which use unsupervised clustering algorithms to find potential illegal users. Then it summarizes the three specific types of fraud, Internet fraud, Ponzi schemes and honeypot contracts, and their "new features" on the Ethereum platform. Further, the relevant research ideas, establishment process and experimental performance evaluation of using machine learning and graph embedding learning algorithms to build detection models are summarized. Finally, the future research direction of Ethereum illegal transaction detections are given.

Key words: Ethereum illegal detection Machine Learning Ponzi scheme honeypot contract Phishing fraud