Salvato in:
Dettagli Bibliografici
Autori principali: Sadat, Souhail Abdelmouaiz, Miliani, Mohamed Yacine Touahria, Hames, Khadidja Hab El, Seba, Hamida, Haddad, Mohammed
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2512.18826
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866917162468245504
author Sadat, Souhail Abdelmouaiz
Miliani, Mohamed Yacine Touahria
Hames, Khadidja Hab El
Seba, Hamida
Haddad, Mohammed
author_facet Sadat, Souhail Abdelmouaiz
Miliani, Mohamed Yacine Touahria
Hames, Khadidja Hab El
Seba, Hamida
Haddad, Mohammed
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2512_18826
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Hyperbolic Graph Embeddings: a Survey and an Evaluation on Anomaly Detection
Sadat, Souhail Abdelmouaiz
Miliani, Mohamed Yacine Touahria
Hames, Khadidja Hab El
Seba, Hamida
Haddad, Mohammed
Machine Learning
Artificial Intelligence
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.
title Hyperbolic Graph Embeddings: a Survey and an Evaluation on Anomaly Detection
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2512.18826