Foundational benchmark: is generated using Unified foundational ontology (UFO)[1].
We claim that a top-level ontology gives a common and recurring shape to both data and queries – our benchmark shows how an index based on OWL representation of UFO implemented invarious triple stores can help making SPARQL evaluation more efficient oversuch data.
The benchmark is used to compare difreent triple stores using generated and existing real datasets the generated dataset is generated by UFO-based Data Generator: which is a RDF triples generator.
It generates data based on a foundational ontology model, the main entities in this models are (Oject, Event, Tropes, Actions).
The generated data is stored in a triple stores and is indexed using UFO index [1].
The users can use the Foundational benchmark to copare different stores w.r.t any data complaint with UFO or By geerting UFO based data with UFO data Generator:
- genetae data using UFO based-model data genetor or use existing UFO-based domain ontology
- persist generatd data into the triple store using JOPA library [3].
- index the data by using UFO-index script.
- instantiate the foundationl patterns to query triple stores.
- run these queries against the triple stores and compare the results.
[1] http://onto.fel.cvut.cz/ontologies/ufo.
[2] AHMAD, J., P. KŘEMEN a M. LEDVINKA. Optimization of Queries Based on Foundational Ontologies. In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). OnTheMove Federated Conferences & Workshops, Valletta, 2018-10-22/2018-10-26. Springer International Publishing, 2018. s. 351-367. ISSN 0302-9743. ISBN 978-3-030-02670-7. DOI 10.1007/978-3-030-02671-4_22.