Problem
Research questions and friction points this paper is trying to address.
Graph Anomaly Detection
Neighbor Information
Spectral Analysis
Graph Neural Networks
Anomaly Signals
NeiGAD: Augmenting Graph Anomaly Detection via Spectral Neighbor Information
🤖 AI Summary
Existing graph anomaly detection methods struggle to explicitly model the interaction between node attributes and local structural neighborhoods. This work proposes NeiGAD, a lightweight, plug-and-play module that, for the first time, introduces spectral graph theory into graph anomaly detection. NeiGAD explicitly encodes local neighbor interactions through the eigenvectors of the adjacency matrix and progressively amplifies anomalous signals to construct discriminative node representations. The module seamlessly integrates with existing models and achieves significant performance gains over state-of-the-art methods across eight real-world datasets, demonstrating the effectiveness of spectral features in enhancing anomaly awareness.
📝 Abstract
Graph anomaly detection (GAD) aims to identify irregular nodes or structures in attributed graphs. Neighbor information, which reflects both structural connectivity and attribute consistency with surrounding nodes, is essential for distinguishing anomalies from normal patterns. Although recent graph neural network (GNN)-based methods incorporate such information through message passing, they often fail to explicitly model its effect or interaction with attributes, limiting detection performance. This work introduces NeiGAD, a novel plug-and-play module that captures neighbor information through spectral graph analysis. Theoretical insights demonstrate that eigenvectors of the adjacency matrix encode local neighbor interactions and progressively amplify anomaly signals. Based on this, NeiGAD selects a compact set of eigenvectors to construct efficient and discriminative representations. Experiments on eight real-world datasets show that NeiGAD consistently improves detection accuracy and outperforms state-of-the-art GAD methods. These results demonstrate the importance of explicit neighbor modeling and the effectiveness of spectral analysis in anomaly detection. Code is available at: https://github.com/huafeihuang/NeiGAD.
Innovation
Methods, ideas, or system contributions that make the work stand out.
spectral graph analysis
graph anomaly detection
neighbor information modeling
eigenvector-based representation
plug-and-play module
🔎 Similar Papers
No similar papers found.
