Improving Generalizability of Graph Anomaly Detection Models Via Data Augmentation

Shuang Zhou, Xiao Huang, Ninghao Liu, Huachi Zhou, Fu Lai Chung, Long Kai Huang

Research output: Journal article publicationJournal articleAcademic researchpeer-review

1 Citation (Scopus)

Abstract

Graph anomaly detection (GAD) has wide applications in real-world networked systems. In many scenarios, people need to identify anomalies on new (sub)graphs, but they may lack labels to train an effective detection model. Since recent semi-supervised GAD methods, which can leverage the available labels as prior knowledge, have achieved superior performance than unsupervised methods, one natural idea is to directly adopt a trained semi-supervised GAD model to the new (sub)graphs for testing. However, we find that existing semi-supervised GAD methods suffer from poor generalization issues, i.e., well-trained models could not perform well on an unseen area (i.e., not accessible in training) of the graph. Motivated by this, we formally define the problem of generalized graph anomaly detection that aims to effectively identify anomalies on both the training-domain graph(s) and the unseen test graph(s). Nevertheless, it is a challenging task since only limited labels are available, and the normal data distribution may differ between training and testing data. Accordingly, we propose a data augmentation method named <italic>AugAN</italic> (<underline>Aug</underline>mentation for <underline>A</underline>nomaly and <underline>N</underline>ormal distributions) to enrich training data and adopt a customized episodic training strategy for learning with the augmented data. Extensive experiments verify the effectiveness of <italic>AugAN</italic> in improving model generalizability.

Original languageEnglish
Pages (from-to)1-14
Number of pages14
JournalIEEE Transactions on Knowledge and Data Engineering
DOIs
Publication statusPublished - May 2023

Keywords

  • Anomaly detection
  • data augmentation
  • Data models
  • Graph anomaly detection
  • Graph neural networks
  • model generalizability
  • Task analysis
  • Testing
  • Training
  • Training data

ASJC Scopus subject areas

  • Information Systems
  • Computer Science Applications
  • Computational Theory and Mathematics

Fingerprint

Dive into the research topics of 'Improving Generalizability of Graph Anomaly Detection Models Via Data Augmentation'. Together they form a unique fingerprint.

Cite this