Rethinking Graph Neural Networks for Anomaly Detection
This work addresses anomaly detection in graph data, which is crucial for applications like fraud detection or network security, but it appears incremental as it builds on existing GNN methods by focusing on spectral analysis.
The paper tackles the problem of graph anomaly detection by analyzing anomalies through the graph spectrum, observing a 'right-shift' phenomenon where spectral energy shifts to higher frequencies, and proposes the Beta Wavelet Graph Neural Network (BWGNN) with localized band-pass filters, demonstrating effectiveness on four large-scale datasets.
Graph Neural Networks (GNNs) are widely applied for graph anomaly detection. As one of the key components for GNN design is to select a tailored spectral filter, we take the first step towards analyzing anomalies via the lens of the graph spectrum. Our crucial observation is the existence of anomalies will lead to the `right-shift' phenomenon, that is, the spectral energy distribution concentrates less on low frequencies and more on high frequencies. This fact motivates us to propose the Beta Wavelet Graph Neural Network (BWGNN). Indeed, BWGNN has spectral and spatial localized band-pass filters to better handle the `right-shift' phenomenon in anomalies. We demonstrate the effectiveness of BWGNN on four large-scale anomaly detection datasets. Our code and data are released at https://github.com/squareRoot3/Rethinking-Anomaly-Detection