Detecting Malicious Accounts in Web3 through Transaction Graph

TL;DR

This paper introduces ScamSweeper, a new framework for detecting malicious accounts in Ethereum's transaction networks, addressing challenges posed by large-scale, temporal, power-law distributed data, and demonstrates superior detection performance.

Contribution

The paper proposes ScamSweeper, a novel detection framework, and provides a large-scale dataset for web3 scam detection, improving over existing methods.

Findings

ScamSweeper outperforms state-of-the-art detection methods.

The dataset includes diverse web3 scam, phishing, and normal accounts.

Large-scale transaction data exhibits power-law distribution with temporal attributes.

Abstract

The web3 applications have recently been growing, especially on the Ethereum platform, starting to become the target of scammers. The web3 scams, imitating the services provided by legitimate platforms, mimic regular activity to deceive users. The current phishing account detection tools utilize graph learning or sampling algorithms to obtain graph features. However, large-scale transaction networks with temporal attributes conform to a power-law distribution, posing challenges in detecting web3 scams. In this paper, we present ScamSweeper, a novel framework to identify web3 scams on Ethereum. Furthermore, we collect a large-scale transaction dataset consisting of web3 scams, phishing, and normal accounts. Our experiments indicate that ScamSweeper exceeds the state-of-the-art in detecting web3 scams.