Abstract:The task of identifying anomalous users on attributed social networks requires the detection of users whose profile attributes and network structure significantly differ from those of the majority of the reference profiles. GNN-based models are well-suited for addressing the challenge of integrating network structure and node attributes into the learning process because they can efficiently incorporate demographic data, activity patterns, and other relevant information. Aggregate operations, such as sum or mean pooling, are utilized by Graph Neural Networks (GNNs) to combine the representations of neighboring nodes within a graph. However, these aggregate operations can cause problems in detecting anomalous nodes. There are two main issues to consider when utilizing aggregate operations in GNNs. Firstly, the presence of anomalous neighboring nodes may affect the representation of normal nodes, leading to false positives. Secondly, anomalous nodes may be overlooked as their representation is flattened during the aggregate operation, leading to false negatives. The proposed approach, AnomEn, is a robust graph neural network developed for anomaly detection. It addresses the challenges of false positives and false negatives using a weighted aggregate mechanism. This mechanism is designed to differentiate between a node's own features and the features of its neighbors by placing greater emphasis on a node's own features and less emphasis on its neighbors' features. The system can preserve the node's original characteristics, whether the node is normal or anomalous. This work proposes not only a robust graph neural network, namely, AnomEn, but also specific anomaly detection structures for nodes and edges. The proposed AnomEn method serves as the encoder in the node and edge anomaly detection architectures and was tested on multiple datasets. Experiments were conducted to validate the effectiveness of the proposed method as a graph neural network encoder. The findings demonstrated the robustness of the proposed method in detecting anomalies. The proposed method outperforms other existing methods in node anomaly detection tasks by 5.63% and edge anomaly detection tasks by 7.87%.

Anomaly-resistant Graph Neural Networks via Neural Architecture Search

Guarding Graph Neural Networks for Unsupervised Graph Anomaly Detection

A Universal Adaptive Algorithm for Graph Anomaly Detection

Robust Graph Neural-Network-Based Encoder for Node and Edge Deep Anomaly Detection on Attributed Networks

Graph Anomaly Detection With Graph Neural Networks: Current Status and Challenges

Accurate Anomaly Detection Leveraging Knowledge-enhanced GAT

Grouped Graph Neural Networks for Anomaly Detection in Time Series

Auto-GNN: Neural Architecture Search of Graph Neural Networks

Rethinking Graph Neural Architecture Search From Message-Passing.

Adversarially Robust Neural Architecture Search for Graph Neural Networks

Adaptive Multi-Information Fusion Model for Enhanced Anomaly Detection

Reinforcement Neighborhood Selection for Unsupervised Graph Anomaly Detection

GAD-NR: Graph Anomaly Detection via Neighborhood Reconstruction

Counterfactual Data Augmentation with Denoising Diffusion for Graph Anomaly Detection

Masked Graph Neural Networks for Unsupervised Anomaly Detection in Multivariate Time Series

Rethinking Graph Neural Network Search from Message-passing

GraphNAS++: Distributed Architecture Search for Graph Neural Networks

Rethinking Graph Neural Networks for Anomaly Detection

Genetic-GNN: Evolutionary architecture search for Graph Neural Networks

Graph Anomaly Detection with Bi-level Optimization