Addressing Graph Anomaly Detection via Causal Edge Separation and Spectrum

2025-08-20

AI Summary

In real-world graphs, anomalous nodes often evade detection by establishing legitimate connections and concealing abnormal associations, thereby forming heterogeneous structures that violate the homophily assumption—leading to degraded performance of conventional graph neural networks (GNNs). Method: We propose CES2-GAD, a novel framework that, for the first time, separates homophilous and heterophilous edges via causal intervention; reveals the spectral energy shift of anomalous nodes toward high frequencies; and designs a hybrid spectral filter to jointly model structural heterogeneity and feature distribution in the spectral domain. Edge separation and spectral analysis are co-optimized under theoretical guidance. Contribution/Results: CES2-GAD establishes a new spectral-domain paradigm for anomaly detection on heterophilous graphs. Extensive experiments on multiple real-world datasets demonstrate an average 12.6% improvement in F1-score over state-of-the-art methods.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Social Networks and Social Media: Fairness and bias in social network and social media analysis
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

In the real world, anomalous entities often add more legitimate connections while hiding direct links with other anomalous entities, leading to heterophilic structures in anomalous networks that most GNN-based techniques fail to address. Several works have been proposed to tackle this issue in the spatial domain. However, these methods overlook the complex relationships between node structure encoding, node features, and their contextual environment and rely on principled guidance, research on solving spectral domain heterophilic problems remains limited. This study analyzes the spectral distribution of nodes with different heterophilic degrees and discovers that the heterophily of anomalous nodes causes the spectral energy to shift from low to high frequencies. To address the above challenges, we propose a spectral neural network CES2-GAD based on causal edge separation for anomaly detection on heterophilic graphs. Firstly, CES2-GAD will separate the original graph into homophilic and heterophilic edges using causal interventions. Subsequently, various hybrid-spectrum filters are used to capture signals from the segmented graphs. Finally, representations from multiple signals are concatenated and input into a classifier to predict anomalies. Extensive experiments with real-world datasets have proven the effectiveness of the method we proposed.

Problem

• Detecting anomalies in heterophilic graphs with shifted spectral energy
• Separating homophilic and heterophilic edges via causal intervention
• Addressing GNN limitations in anomalous network structures

Innovation

• Causal edge separation for heterophilic graphs
• Hybrid-spectrum filters capture segmented signals
• Concatenated representations input to anomaly classifier