DGraph: A Large-Scale Financial Dataset for Graph Anomaly Detection

TL;DR

This paper introduces DGraph, a large-scale, real-world financial graph dataset designed to enhance graph anomaly detection research by providing diverse, dynamic, and annotated data for better understanding and detection of anomalies.

Contribution

The paper presents DGraph, a comprehensive large-scale financial graph dataset that addresses limitations of existing datasets and facilitates advanced GAD research.

Findings

DGraph contains 3 million nodes and 4 million dynamic edges.

Anomalous nodes differ significantly from normal nodes in structure and dynamics.

Unlabeled nodes are important for detecting fraudsters.

Abstract

Graph Anomaly Detection (GAD) has recently become a hot research spot due to its practicability and theoretical value. Since GAD emphasizes the application and the rarity of anomalous samples, enriching the varieties of its datasets is fundamental work. Thus, this paper present DGraph, a real-world dynamic graph in the finance domain. DGraph overcomes many limitations of current GAD datasets. It contains about 3M nodes, 4M dynamic edges, and 1M ground-truth nodes. We provide a comprehensive observation of DGraph, revealing that anomalous nodes and normal nodes generally have different structures, neighbor distribution, and temporal dynamics. Moreover, it suggests that unlabeled nodes are also essential for detecting fraudsters. Furthermore, we conduct extensive experiments on DGraph. Observation and experiments demonstrate that DGraph is propulsive to advance GAD research and enable in-depth exploration of anomalous nodes.