FGC-Comp: Adaptive Neighbor-Grouped Attribute Completion for Graph-based Anomaly Detection

TL;DR

This work tackles missing and adversarially obscured node attributes in graph-based anomaly detection by introducing FGC-Comp, a lightweight, classifier-agnostic attribute completion module. It partitions neighbors into fraud/benign/unknown groups and uses a node-conditioned gate to adapt unknowns via a convex mixture of two group transforms, with residual fusion feeding a lightweight predictor. Trained end-to-end with binary cross-entropy on standard encoders, FGC-Comp achieves robust improvements on YelpChi and Amazon datasets with negligible overhead. The results demonstrate improved stability and reliability under missing attributes, highlighting deployment-friendly design without heavy architectural changes.

Abstract

Graph-based Anomaly Detection models have gained widespread adoption in recent years, identifying suspicious nodes by aggregating neighborhood information. However, most existing studies overlook the pervasive issues of missing and adversarially obscured node attributes, which can undermine aggregation stability and prediction reliability. To mitigate this, we propose FGC-Comp, a lightweight, classifier-agnostic, and deployment-friendly attribute completion module-designed to enhance neighborhood aggregation under incomplete attributes. We partition each node's neighbors into three label-based groups, apply group-specific transforms to the labeled groups while a node-conditioned gate handles unknowns, fuse messages via residual connections, and train end-to-end with a binary classification objective to improve aggregation stability and prediction reliability under missing attributes. Experiments on two real-world fraud datasets validate the effectiveness of the approach with negligible computational overhead.

Figures (2)

• Figure 1: FGC-Comp overview.
• Figure 2: Amazon & YelpChi: Recall and AUC vs. Feature dropout ratio.