Anomaly Detection in Graph Structured Data: A Survey
Prabin B Lamichhane, William Eberle
TL;DR
This survey addresses the problem of anomaly detection in graph-structured data by presenting a taxonomy that organizes methods according to anomalous components, graph data types (static vs dynamic), and technique families. It synthesizes probabilistic/statistical, matrix/tensor, distance-based, and graph neural network approaches, highlighting their applicability to nodes, edges, and subgraphs across diverse domains. The work also catalogues datasets and practical challenges, including dynamics, scalability, interpretability, and robustness to adversarial perturbations, and it outlines future research directions. Overall, the paper provides a comprehensive, actionable roadmap for researchers and practitioners aiming to develop scalable, interpretable graph-based anomaly detection solutions with impact in security, finance, and large-scale networked systems.
Abstract
Real-world graphs are complex to process for performing effective analysis, such as anomaly detection. However, recently, there have been several research efforts addressing the issues surrounding graph-based anomaly detection. In this paper, we discuss a comprehensive overview of anomaly detection techniques on graph data. We also discuss the various application domains which use those anomaly detection techniques. We present a new taxonomy that categorizes the different state-of-the-art anomaly detection methods based on assumptions and techniques. Within each category, we discuss the fundamental research ideas that have been done to improve anomaly detection. We further discuss the advantages and disadvantages of current anomaly detection techniques. Finally, we present potential future research directions in anomaly detection on graph-structured data.