The RDF2vec family of knowledge graph embedding methods : An experimental evaluation of RDF2vec variants and their capabilitie

DOI:	https://doi.org/10.3233/SW-233514
URL:	https://content.iospress.com/articles/semantic-web...
Dokumenttyp:	Zeitschriftenartikel
Erscheinungsjahr:	2024
Titel einer Zeitschrift oder einer Reihe:	Semantic Web
Band/Volume:	15
Heft/Issue:	3
Seitenbereich:	845-876
Ort der Veröffentlichung:	Amsterdam
Verlag:	IOS Press
ISSN:	1570-0844 , 2210-4968
Sprache der Veröffentlichung:	Englisch
Einrichtung:	Fakultät für Wirtschaftsinformatik und Wirtschaftsmathematik > Data Science (Paulheim 2018-)
Fachgebiet:	004 Informatik
Freie Schlagwörter (Englisch):	RDF2vec , knowledge graph embedding , representation learning , embedding evaluation
Abstract:	Knowledge graph embeddings represent a group of machine learning techniques which project entities and relations of a knowledge graph to continuous vector spaces. RDF2vec is a scalable embedding approach rooted in the combination of random walks with a language model. It has been successfully used in various applications. Recently, multiple variants to the RDF2vec approach have been proposed, introducing variations both on the walk generation and on the language modeling side. The combination of those different approaches has lead to an increasing family of RDF2vec variants. In this paper, we evaluate a total of twelve RDF2vec variants on a comprehensive set of benchmark models, and compare them to seven existing knowledge graph embedding methods from the family of link prediction approaches. Besides the established GEval benchmark introducing various downstream machine learning tasks on the DBpedia knowledge graph, we also use the new DLCC (Description Logic Class Constructors) benchmark consisting of two gold standards, one based on DBpedia, and one based on synthetically generated graphs. The latter allows for analyzing which ontological patterns in a knowledge graph can actually be learned by different embedding. With this evaluation, we observe that certain tailored RDF2vec variants can lead to improved performance on different downstream tasks, given the nature of the underlying problem, and that they, in particular, have a different behavior in modeling similarity and relatedness. The findings can be used to provide guidance in selecting a particular RDF2vec method for a given task.