Temporal-Aware Graph Attention Network for Cryptocurrency Transaction Fraud Detection

TL;DR

This paper introduces ATGAT, a novel temporal-aware graph attention network that significantly improves cryptocurrency transaction fraud detection by capturing complex temporal and structural dependencies.

Contribution

It proposes a new model with advanced temporal embedding, triple attention mechanism, and weighted loss to enhance detection performance over traditional methods.

Findings

Achieves an AUC of 0.9130 on Elliptic++ dataset.

Outperforms traditional methods like XGBoost, GCN, and GAT.

Demonstrates the effectiveness of temporal awareness and triple attention in fraud detection.

Abstract

Cryptocurrency transaction fraud detection faces the dual challenges of increasingly complex transaction patterns and severe class imbalance. Traditional methods rely on manual feature engineering and struggle to capture temporal and structural dependencies in transaction networks. This paper proposes an Augmented Temporal-aware Graph Attention Network (ATGAT) that enhances detection performance through three modules: (1) designing an advanced temporal embedding module that fuses multi-scale time difference features with periodic position encoding; (2) constructing a temporal-aware triple attention mechanism that jointly optimizes structural, temporal, and global context attention; (3) employing weighted BCE loss to address class imbalance. Experiments on the Elliptic++ cryptocurrency dataset demonstrate that ATGAT achieves an AUC of 0.9130, representing a 9.2% improvement over the best traditional method XGBoost, 12.0% over GCN, and 10.0% over standard GAT. This method not only validates the enhancement effect of temporal awareness and triple attention mechanisms on graph neural networks, but also provides financial institutions with more reliable fraud detection tools, with its design principles generalizable to other temporal graph anomaly detection tasks.