用双曲空间提升图异常检测，效果显著优于传统方法。
Hyperbolic Graph Embeddings: a Survey and an Evaluation on Anomaly Detection
- 采用双曲几何建模复杂图结构，更契合层级关系
- P-VAE在Elliptic数据集上达94%的F1分数，表现最优
- 适合处理具有层次特性的网络数据，如社交或金融图
本综述系统评估了双曲图嵌入模型在异常检测中的表现，揭示其在捕捉复杂结构方面优于欧几里得方法。实验对比了HGCAE、P-VAE和HGCN等模型，结果显示，P-VAE在Elliptic数据集上取得94%的F1分数，HGCAE在Cora数据集上达到80%的性能；而欧几里得方法如DOMINANT和GraphSage在复杂数据上表现不佳。研究强调双曲空间在异常检测中的潜力，并开源了相关工具库，以推动该领域进一步发展。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。