Designing Hierarchies for Optimal Hyperbolic Embedding

Melika Ayoughi, Max van Spengler, Pascal Mettes, Paul Groth

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Hyperbolic geometry has shown to be highly effective for embedding hierarchical data structures. As such, machine learning in hyperbolic space is rapidly gaining traction across a wide range of disciplines, from recommender systems and graph networks to biological systems and computer vision. The performance of hyperbolic learning commonly depends on the hierarchical information used as input or supervision. Given that knowledge graphs and ontologies are common sources of such hierarchies, this paper aims to guide ontology designers in designing hierarchies for use in these learning algorithms. Using widely employed measures of embedding quality with extensive experiments, we find that hierarchies are best suited for hyperbolic embeddings when they are wide, and single inheritance, independent of the hierarchy size and imbalance.

Original languageEnglish
Title of host publicationThe Semantic Web - 22nd European Semantic Web Conference, ESWC 2025, Proceedings
EditorsEdward Curry, Maribel Acosta, Maria Poveda-Villalón, Marieke van Erp, Adegboyega Ojo, Katja Hose, Cogan Shimizu, Pasquale Lisena
PublisherSpringer Science and Business Media Deutschland GmbH
Pages362-382
Number of pages21
ISBN (Print)9783031945748
DOIs
StatePublished - 2025
Externally publishedYes
Event22nd European Semantic Web Conference, ESWC 2025 - Portoroz, Slovenia
Duration: Jun 1 2025Jun 5 2025

Publication series

NameLecture Notes in Computer Science
Volume15718 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference22nd European Semantic Web Conference, ESWC 2025
Country/TerritorySlovenia
CityPortoroz
Period06/1/2506/5/25

Keywords

  • Hyperbolic Learning
  • Machine Learning
  • Ontology Design

Fingerprint

Dive into the research topics of 'Designing Hierarchies for Optimal Hyperbolic Embedding'. Together they form a unique fingerprint.

Cite this