aiqbench analysis
This paper introduces a toolkit and eleven datasets for knowledge graph refinement that keep the full OWL schema, not just ground facts, so neural-symbolic methods can be evaluated on expressive schemas. Its practical value lies in reproducible benchmarks and direct export to PyTorch and PyKEEN. The reported validation is preliminary, so readers should check how well the schema-aware gains hold up on established tasks before drawing conclusions.
JediKG: Return of the Schema
arXiv:2602.14795v2 Announce Type: replace-cross Abstract: Benchmark datasets for Knowledge Graph Refinement typically contain only ground facts, discarding the schema available in the source knowledge graph. This limits the evaluation of Neural-Symbolic methods that leverage expressive schemas. We present JEDIKG the first resource that systematically addresses these gaps through two components: JDEX a general configurable tool for extracting complete, portable, modular dataset from any RDF/OWL knowledge graph; and JDSET, a curated suite of eleven datasets with varying expressivity, volume, and domain. JDEX preserves full schema expressivity, employs standard OWL reasoning to detect and correct inconsistencies and materialize implicit knowledge, applies schema modularization, and provides utilities for integration with mainstream machine learning libraries such as PyTorch and PyKEEN. We validated the completeness and portability of the datasets across several dimensions and preliminarily proved the utility of JEDIKG in experimental workflows combining data-driven methods with schema-driven reasoning.
Abstract:Benchmark datasets for Knowledge Graph Refinement typically contain only ground facts, discarding the schema available in the source knowledge graph. This limits the evaluation of Neural-Symbolic methods that leverage expressive schemas. We present JEDIKG the first resource that systematically addresses these gaps through two components: JDEX a general configurable tool for extracting complete, portable, modular dataset from any RDF/OWL knowledge graph; and JDSET, a curated suite of eleven datasets with varying expressivity, volume, and domain. JDEX preserves full schema expressivity, employs standard OWL reasoning to detect and correct inconsistencies and materialize implicit knowledge, applies schema modularization, and provides utilities for integration with mainstream machine learning libraries such as PyTorch and PyKEEN. We validated the completeness and portability of the datasets across several dimensions and preliminarily proved the utility of JEDIKG in experimental workflows combining data-driven methods with schema-driven reasoning.
This preprint has not undergone peer review (when applicable) or any post-submission improvements or corrections. The Version of Record of this contribution is published in The Semantic Web - ISWC 2026, and is available online at this https URL
Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
arXiv:2602.14795 [cs.AI]
(or arXiv:2602.14795v2 [cs.AI] for this version)
In: The Semantic Web - ISWC 2026. LNCS, vol 17127, Springer, Cham
https://doi.org/10.1007/978-3-032-42026-8_21 DOI(s) linking to related resources
Full text under the publisher's open licence · 210 words