Home // SEMAPRO 2015, The Ninth International Conference on Advances in Semantic Processing // View article
Authors:
Raoul Kwuimi
Jean Vincent Fonou-Dombeu
Keywords: Jena API; Sesame; SPARQL; Ontology Storage; Relational Databases .
Abstract:
Mapping ontologies to relational databases is an active research topic in Semantic Web. Therefore, several platforms have been developed to enable the storage and query of ontologies in relational databases. However, only a few studies have empirically measured and compared their performances in terms of speed and scalability. In this paper, two popular database-based ontologies stores, namely, Jena API and Sesame are used to load and query five selected ontologies of different sizes into MySQL relational database. Various metrics including (1) the loading times of ontologies into the relational databases, (2) the response times of SPARQL queries executed on the stored ontologies databases and (3) the sizes of the ontologies databases are used to measure and compare the performance of the two Semantic Web platforms. Experiments show that (1) both platforms are scalable and could successfully parse, load and query ontologies of different formats (OWL/RDF) and sizes into relational databases, (2) Jena API performs faster with small size ontologies, whereas, Sesame is more efficient with bigger size ontologies with regards to loading of ontologies into relational databases, (3) Sesame provides quicker responses to SPARQL queries compared to Jena API and (4) the disk space required to store the resulting ontologies databases in both platforms are proportional to the initial sizes of the ontologies and is higher in Jena API than in Sesame.
Pages: 6 to 12
Copyright: Copyright (c) IARIA, 2015
Publication date: July 19, 2015
Published in: conference
ISSN: 2308-4510
ISBN: 978-1-61208-420-6
Location: Nice, France
Dates: from July 19, 2015 to July 24, 2015