Cross-Domain Graph Level Anomaly Detection

IEEE Transactions on Knowledge and Data Engineering, vol. 36, pp. 7839–7850

Abstract

Existing graph level anomaly detection methods are predominantly unsupervised due to high costs for obtaining labels, yielding sub-optimal detection accuracy when compared to supervised methods. Moreover, they heavily rely on the assumption that the training data exclusively consists of normal graphs. Hence, even the presence of a few anomalous graphs can lead to substantial performance degradation. To alleviate these problems, we propose across-domain graph level anomaly detection method, aiming to identify anomalous graphs from a set of unlabeled graphs (target domain) by using easily accessible normal graphs from a different but related domain (source domain). Our method consists of four components: a feature extractor that preserves semantic and topological information of individual graphs while incorporating the distance between different graphs; an adversarial domain classifier to make graph level representations domain-invariant; a one-class classifier to exploit label information in the source domain; and a class aligner to align classes from both domains based on pseudolabels. Experiments on seven benchmark datasets show that the proposed method largely outperforms state-of-the-art methods.

Authors 4

1. Zhong Li Aachen

   Leiden University

   Affiliation as printed

   LIACS, Leiden University, Leiden, The Netherlands

2. Sheng Liang

   Ludwig-Maximilians-Universität München

   Affiliation as printed

   CIS, LMU Munich, München, Germany

3. Jiayang Shi Aachen

   Leiden University

   Affiliation as printed

   LIACS, Leiden University, Leiden, The Netherlands

4. Matthijs van Leeuwen Aachen

   Leiden University

   Affiliation as printed

   LIACS, Leiden University, Leiden, The Netherlands