SPARQLing Datalog for Rule-Based Reasoning over Large Knowledge Graphs (Extended Abstract)

Alex IvlievAlex Ivliev, Markus KrötzschMarkus Krötzsch, Maximilian MarxMaximilian Marx

Alex Ivliev, Markus Krötzsch, Maximilian Marx
SPARQLing Datalog for Rule-Based Reasoning over Large Knowledge Graphs (Extended Abstract)
In Marco Calautti, Matthias Lanzinger, eds., Proceedings of the 6th International Workshop on the Resurgence of Datalog in Academia and Industry (Datalog-2.0 2026), CEUR Workshop Proceedings, to appear. CEUR-WS.org

Details
Bibtex

KurzfassungAbstract
Rule languages such as Datalog are well-suited for reasoning over graph-structured data, typically represented as RDF triples. However, the largest knowledge graphs cannot be processed by existing in-memory rule engines. We have recently proposed an integration of rule engines with external RDF stores in which relevant data is fetched during reasoning through automatically generated SPARQL queries. Users write plain Datalog programs over RDF triples, without any knowledge of SPARQL, while the reasoner constructs and optimises the required queries. The optimisations adapt established techniques from logic programming, such as semi-naive evaluation, magic sets, and static filtering. In this extended abstract, we summarise our approach, implemented in the rule engine Nemo, and the empirical evaluation in which complex rule sets are executed against the public SPARQL services of Wikidata and DBLP.
Projekt:Project: CPEC, SECAI, ScaDS.AI
Forschungsgruppe:Research Group: Wissensbasierte SystemeKnowledge-Based Systems

@inproceedings{IKM2026,
  author    = {Alex Ivliev and Markus Kr{\"{o}}tzsch and Maximilian Marx},
  title     = {SPARQLing Datalog for Rule-Based Reasoning over Large Knowledge
               Graphs (Extended Abstract)},
  editor    = {Marco Calautti and Matthias Lanzinger},
  booktitle = {Proceedings of the 6th International Workshop on the Resurgence
               of Datalog in Academia and Industry (Datalog-2.0 2026)},
  series    = {CEUR Workshop Proceedings},
  publisher = {CEUR-WS.org},
  year      = {2026}
}