analiza aiqbench
This survey addresses the fragmentation in heterogeneous Graph Neural Network (HGNN) research for cybersecurity by introducing a taxonomy that classifies methods based on anomaly types and graph dynamics. It critically evaluates existing benchmark datasets and metrics while highlighting the lack of standardized comparisons that currently hinder practical deployment. The work provides a structured foundation for future research aimed at developing scalable and interpretable solutions for detecting insider threats and coordinated attacks.
A Survey of Heterogeneous Graph Neural Networks for Cybersecurity Anomaly Detection
arXiv:2510.26307v3 Announce Type: replace-cross Abstract: Anomaly detection is a critical task in cybersecurity, where identifying insider threats, access violations, and coordinated attacks is essential for ensuring system resilience. Graph-based approaches have become increasingly important for modeling entity interactions, yet most rely on homogeneous and static structures, which limits their ability to capture the heterogeneity and temporal evolution of real-world environments. Heterogeneous Graph Neural Networks (HGNNs) have emerged as a promising paradigm for anomaly detection by incorporating type-aware transformations and relation-sensitive aggregation, enabling more expressive modeling of complex cyber data. However, current research on HGNN-based anomaly detection remains fragmented, with diverse modeling strategies, limited comparative evaluation, and an absence of standardized benchmarks. To address this gap, we provide a comprehensive survey of HGNN-based anomaly detection methods in cybersecurity. We introduce a taxonomy that classifies approaches by anomaly type and graph dynamics, analyze representative models, and map them to key cybersecurity applications. We also review commonly used benchmark datasets and evaluation metrics, highlighting their strengths and limitations. Finally, we identify key open challenges related to modeling, data, and deployment, and outline promising directions for future research. This survey aims to establish a structured foundation for advancing HGNN-based anomaly detection toward scalable, interpretable, and practically deployable solutions.
Abstract:Anomaly detection is a critical task in cybersecurity, where identifying insider threats, access violations, and coordinated attacks is essential for ensuring system resilience. Graph-based approaches have become increasingly important for modeling entity interactions, yet most rely on homogeneous and static structures, which limits their ability to capture the heterogeneity and temporal evolution of real-world environments. Heterogeneous Graph Neural Networks (HGNNs) have emerged as a promising paradigm for anomaly detection by incorporating type-aware transformations and relation-sensitive aggregation, enabling more expressive modeling of complex cyber data. However, current research on HGNN-based anomaly detection remains fragmented, with diverse modeling strategies, limited comparative evaluation, and an absence of standardized benchmarks. To address this gap, we provide a comprehensive survey of HGNN-based anomaly detection methods in cybersecurity. We introduce a taxonomy that classifies approaches by anomaly type and graph dynamics, analyze representative models, and map them to key cybersecurity applications. We also review commonly used benchmark datasets and evaluation metrics, highlighting their strengths and limitations. Finally, we identify key open challenges related to modeling, data, and deployment, and outline promising directions for future research. This survey aims to establish a structured foundation for advancing HGNN-based anomaly detection toward scalable, interpretable, and practically deployable solutions.
23 pages, 7 figures, and 97 references. Accepted by the Journal of Computer Security
Cryptography and Security (cs.CR); Machine Learning (cs.LG)
arXiv:2510.26307 [cs.CR]
(or arXiv:2510.26307v3 [cs.CR] for this version)
Text integral sub licența deschisă a publicației · 228 cuvinte
Subiecte conexe
Chi-Square Wavelet Graph Neural Networks for Heterogeneous Graph Anomaly Detection
arXiv · 23 iul. 2026
ServiceNow AI Control Tower: Selling a penthouse while building the frame
SiliconANGLE · 17 oct. 2026
TH-GNN: Heterogeneous Temporal Graph Neural Networks for LLM-Agent Shilling Attack Detection
arXiv · 24 aug. 2026
Diffusion-Induced Spatial Attention Overlapping Community Detection
arXiv · 23 sept. 2026