How to Use Graph Data in the Wild to Help Graph Anomaly Detection?

TL;DR

This paper introduces Wild-GAD, a framework leveraging diverse external graph data to improve anomaly detection in graph-structured data, addressing challenges like label scarcity and anomaly variability.

Contribution

It proposes a novel approach using external graph data and a unified database to enhance anomaly detection accuracy in various domains.

Findings

18% average AUCROC improvement over baselines

32% average AUCPR improvement over baselines

Effective selection of external data based on representativity and diversity

Abstract

In recent years, graph anomaly detection has found extensive applications in various domains such as social, financial, and communication networks. However, anomalies in graph-structured data present unique challenges, including label scarcity, ill-defined anomalies, and varying anomaly types, making supervised or semi-supervised methods unreliable. Researchers often adopt unsupervised approaches to address these challenges, assuming that anomalies deviate significantly from the normal data distribution. Yet, when the available data is insufficient, capturing the normal distribution accurately and comprehensively becomes difficult. To overcome this limitation, we propose to utilize external graph data (i.e., graph data in the wild) to help anomaly detection tasks. This naturally raises the question: How can we use external data to help graph anomaly detection tasks? To answer this question, we propose a framework called Wild-GAD. It is built upon a unified database, UniWildGraph, which comprises a large and diverse collection of graph data with broad domain coverage, ample data volume, and a unified feature space. Further, we develop selection criteria based on representativity and diversity to identify the most suitable external data for anomaly detection task. Extensive experiments on six real-world datasets demonstrate the effectiveness of Wild-GAD. Compared to the baseline methods, our framework has an average 18% AUCROC and 32% AUCPR improvement over the best-competing methods.