AutoRDF2GML: Facilitating RDF Integration in Graph Machine Learning

  • 2024-07-26 14:44:06
  • Michael Färber, David Lamprecht, Yuni Susanti
  • 0

Abstract

In this paper, we introduce AutoRDF2GML, a framework designed to convert RDFdata into data representations tailored for graph machine learning tasks.AutoRDF2GML enables, for the first time, the creation of both content-basedfeatures -- i.e., features based on RDF datatype properties -- andtopology-based features -- i.e., features based on RDF object properties.Characterized by automated feature extraction, AutoRDF2GML makes it possibleeven for users less familiar with RDF and SPARQL to generate datarepresentations ready for graph machine learning tasks, such as linkprediction, node classification, and graph classification. Furthermore, wepresent four new benchmark datasets for graph machine learning, created fromlarge RDF knowledge graphs using our framework. These datasets serve asvaluable resources for evaluating graph machine learning approaches, such asgraph neural networks. Overall, our framework effectively bridges the gapbetween the Graph Machine Learning and Semantic Web communities, paving the wayfor RDF-based machine learning applications.

 

Quick Read (beta)

loading the full paper ...