Hyperbolic Graph Embeddings: a Survey and an Evaluation on Anomaly Detection
By: Souhail Abdelmouaiz Sadat , Mohamed Yacine Touahria Miliani , Khadidja Hab El Hames and more
This survey reviews hyperbolic graph embedding models, and evaluate them on anomaly detection, highlighting their advantages over Euclidean methods in capturing complex structures. Evaluating models like \textit{HGCAE}, \textit{\(\mathcal{P}\)-VAE}, and \textit{HGCN} demonstrates high performance, with \textit{\(\mathcal{P}\)-VAE} achieving an F1-score of 94\% on the \textit{Elliptic} dataset and \textit{HGCAE} scoring 80\% on \textit{Cora}. In contrast, Euclidean methods like \textit{DOMINANT} and \textit{GraphSage} struggle with complex data. The study emphasizes the potential of hyperbolic spaces for improving anomaly detection, and provides an open-source library to foster further research in this field.
Similar Papers
Combining Euclidean and Hyperbolic Representations for Node-level Anomaly Detection
Machine Learning (CS)
Finds strange patterns in connected data.
HyperComplEx: Adaptive Multi-Space Knowledge Graph Embeddings
Artificial Intelligence
Helps computers understand complex information better.
sHGCN: Simplified hyperbolic graph convolutional neural networks
Machine Learning (CS)
Makes computers understand complex data much faster.