Fraud Detection in Cryptocurrency Markets with Spatio-Temporal Graph Neural Networks

• URL: http://arxiv.org/abs/2604.24590v1
• Date: Mon, 27 Apr 2026 15:12:11 GMT
• Title: Fraud Detection in Cryptocurrency Markets with Spatio-Temporal Graph Neural Networks
• Abstract summary: We propose three graph construction methods that rely on aggregated hourly market data.<n>We evaluate our methodology on a real-world dataset comprised of pump-and-dump schemes in cryptocurrency markets.<n>Our work highlights that learned market connectivity provides substantial gains for detecting coordinated market manipulation schemes.
• Score: 0.9332987715848716
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Technological advancements in cryptocurrency markets have increased accessibility for investors, but concurrently exposed them to the risks of market manipulations. Existing fraud detection mechanisms typically rely on machine learning methods that treat each financial asset (i.e., token) and its related transactions independently. However, market manipulation strategies are rarely isolated events, but are rather characterized by coordination, repetition, and frequent transfers among related assets. This suggests that relational structure constitutes an integral component of the signal and can be effectively represented through graphical means. In this paper, we propose three graph construction methods that rely on aggregated hourly market data. The proposed graphs are processed by a unified spatio-temporal Graph Neural Network (GNN) architecture that combines attention-based spatial aggregation with temporal Transformer encoding. We evaluate our methodology on a real-world dataset comprised of pump-and-dump schemes in cryptocurrency markets, spanning a period of over three years. Our comparative results showcase that our graph-based models achieve significant improvements over standard machine learning baselines in detecting anomalous events. Our work highlights that learned market connectivity provides substantial gains for detecting coordinated market manipulation schemes.

Related papers

• A Sustainable AI Economy Needs Data Deals That Work for Generators [56.949279542190084]
  We argue that the machine learning value chain is structurally unsustainable due to an economic data processing inequality.<n>We analyze 73 public data deals and show that the majority of value accrues to aggregators.<n>We propose an Equitable Data-Value Exchange Framework to enable a minimal market that benefits all participants.
  arXiv  Detail & Related papers  (2026-01-15T01:05:48Z)
• DeXposure: A Dataset and Benchmarks for Inter-protocol Credit Exposure in Decentralized Financial Networks [21.810918962434712]
  The DeXposure dataset is the first large-scale dataset for inter-protocol credit exposure in decentralized financial networks.<n>It covers global markets of 43.7 million entries across 4.3 thousand protocols, 602 blockchains, and 24.3 thousand tokens from 2020 to 2025.<n>The DeXposure dataset and code have been released publicly.
  arXiv  Detail & Related papers  (2025-11-27T10:44:14Z)
• News-Aware Direct Reinforcement Trading for Financial Markets [4.651395723728895]
  In this work, we use the news sentiment scores derived from large language models, together with raw price and volume data, as observable inputs for reinforcement learning.<n>These inputs are processed by sequence models such as recurrent neural networks or Transformers to make end-to-end trading decisions.<n>The results demonstrate that our news-aware approach, which does not depend on handcrafted features or manually designed rules, can achieve performance superior to market benchmarks.
  arXiv  Detail & Related papers  (2025-10-22T02:17:03Z)
• A Graph Machine Learning Approach for Detecting Topological Patterns in Transactional Graphs [0.9199488958939334]
  The rise of digital ecosystems has exposed the financial sector to evolving abuse and criminal tactics.<n>Traditional rule-based systems lack the adaptability needed to detect sophisticated or coordinated criminal behaviors.<n>We propose an approach that integrates graph machine learning and network analysis to improve the detection of well-known topological patterns.
  arXiv  Detail & Related papers  (2025-09-16T06:43:11Z)
• ProteuS: A Generative Approach for Simulating Concept Drift in Financial Markets [44.76567557906836]
  A fundamental problem in developing and validating adaptive algorithms is the lack of a ground truth in real-world financial data.<n>This paper introduces a novel framework, named ProteuS, for generating semi-synthetic financial time series with pre-defined structural breaks.<n>An analysis of the generated data confirms the complexity of the task, revealing significant overlap between the different market states.
  arXiv  Detail & Related papers  (2025-08-30T21:01:47Z)
• A Network Simulation of OTC Markets with Multiple Agents [3.8944986367855963]
  We present a novel approach to simulating an over-the-counter (OTC) financial market in which trades are intermediated solely by market makers. We show that our network-based model can lend insights into the effect of market-structure on price-action.
  arXiv  Detail & Related papers  (2024-05-03T20:45:00Z)
• Enhancing Smart Contract Security Analysis with Execution Property Graphs [48.31617821205042]
  We introduce Clue, a dynamic analysis framework specifically designed for a runtime virtual machine.<n>Clue captures critical information during contract executions, employing a novel graph-based representation, the Execution Property Graph.<n> evaluation results reveal Clue's superior performance with high true positive rates and low false positive rates, outperforming state-of-the-art tools.
  arXiv  Detail & Related papers  (2023-05-23T13:16:42Z)
• Temporal and Heterogeneous Graph Neural Network for Financial Time Series Prediction [14.056579711850578]
  We propose a temporal and heterogeneous graph neural network-based (THGNN) approach to learn the dynamic relations among price movements in financial time series. We conduct extensive experiments on the stock market in the United States and China.
  arXiv  Detail & Related papers  (2023-05-09T11:17:46Z)
• Augmented Bilinear Network for Incremental Multi-Stock Time-Series Classification [83.23129279407271]
  We propose a method to efficiently retain the knowledge available in a neural network pre-trained on a set of securities. In our method, the prior knowledge encoded in a pre-trained neural network is maintained by keeping existing connections fixed. This knowledge is adjusted for the new securities by a set of augmented connections, which are optimized using the new data.
  arXiv  Detail & Related papers  (2022-07-23T18:54:10Z)
• Temporal-Relational Hypergraph Tri-Attention Networks for Stock Trend Prediction [45.74513775015998]
  We present a collaborative temporal-relational modeling framework for end-to-end stock trend prediction. A novel hypergraph tri-attention network (HGTAN) is proposed to augment the hypergraph convolutional networks. In this manner, HGTAN adaptively determines the importance of nodes, hyperedges, and hypergraphs during the information propagation among stocks.
  arXiv  Detail & Related papers  (2021-07-22T02:16:09Z)
• Taking Over the Stock Market: Adversarial Perturbations Against Algorithmic Traders [47.32228513808444]
  We present a realistic scenario in which an attacker influences algorithmic trading systems by using adversarial learning techniques. We show that when added to the input stream, our perturbation can fool the trading algorithms at future unseen data points.
  arXiv  Detail & Related papers  (2020-10-19T06:28:05Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.