\textsc{Perseus}: Tracing the Masterminds Behind Cryptocurrency Pump-and-Dump Schemes

TL;DR

Perseus tackles the root cause of crypto pump-and-dump by tracing masterminds through real-time osn and market data, building temporal diffusion graphs, and applying GNNs to classify spreaders as masterminds or accomplices. The system combines a real-time fetcher, a temporal graph generator, and a mastermind detector based on Graph Attention Networks and GraphSAGE, achieving superior precision and F1 scores while enabling fast training and inference. Real-world deployment in 2024 demonstrated detection of 438 masterminds across 322 cryptos and revealed substantial market impact, with regulators gaining actionable risk explanations. The work advances financial forensics in crypto by providing an interpretable, scalable framework for identifying upstream manipulators and informing oversight strategies.

Abstract

Masterminds are entities organizing, coordinating, and orchestrating cryptocurrency pump-and-dump schemes, a form of trade-based manipulation undermining market integrity and causing financial losses for unwitting investors. Previous research detects pump-and-dump activities in the market, predicts the target cryptocurrency, and examines investors and \ac{osn} entities. However, these solutions do not address the root cause of the problem. There is a critical gap in identifying and tracing the masterminds involved in these schemes. In this research, we develop a detection system \textsc{Perseus}, which collects real-time data from the \acs{osn} and cryptocurrency markets. \textsc{Perseus} then constructs temporal attributed graphs that preserve the direction of information diffusion and the structure of the community while leveraging \ac{gnn} to identify the masterminds behind pump-and-dump activities. Our design of \textsc{Perseus} leads to higher F1 scores and precision than the \ac{sota} fraud detection method, achieving fast training and inferring speeds. Deployed in the real world from February 16 to October 9 2024, \textsc{Perseus} successfully detects $438$ masterminds who are efficient in the pump-and-dump information diffusion networks. \textsc{Perseus} provides regulators with an explanation of the risks of masterminds and oversight capabilities to mitigate the pump-and-dump schemes of cryptocurrency.

Figures (12)

• Figure 1: Proportion of Pumps Over Time.
• Figure 2: Kernel Density of Pumps per Channel.
• Figure 4: The crowd-pump message coordinates investors to purchase the cryptocurrency when the price is within the entry prices and sell them at the target prices.
• Figure 5: The market reaction to the crowd-pump message in \ref{['subfig:crowd_pump message']}.
• Figure 7: Perseus consists of three components: real-time fetcher, temporal attributed graph generator, and mastermind detector. The real-time fetcher collects data from osn and the cryptocurrency market, which is processed by temporal attributed graph generators to create information diffusion graphs that serve as input for the mastermind detector to identify masterminds. The real-time fetcher and temporal attributed graph generator concurrently process data from osn and cryptocurrency markets.