Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Ontology Materialization by Abstraction Refinement in Horn SHOIF
Authors: Birte Glimm, Yevgeny Kazakov, Trung-Kien Tran
AAAI 2017 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | An empirical evaluation demonstrates that, despite the new features, the abstractions are still significantly smaller than the original ontologies and the materialization can be computed efficiently. We implemented a prototype system Orar for full materialization of Horn SHOIF ontologies, evaluated Orar on popular ontologies, and compared it with other reasoners. Table 3 presents detailed information about the test ontologies and the experimental results. |
| Researcher Affiliation | Academia | Birte Glimm and Yevgeny Kazakov and Trung-Kien Tran Institute of Artificial Intelligence, University of Ulm, Germany <first name>.<last name>@uni-ulm.de |
| Pseudocode | No | The general algorithm for ontology reasoning using the abstraction refinement method can be summarized as follows: 1. Build a suitable abstraction of the original ontology; 2. Compute the entailments from the abstraction using a reasoner and transfer them to the original ontology using homomorphisms (Lemma 1); 3. Compute the deductive closure of the original ontology using some (light-weight) rules; 4. Repeat from Step 1 until no new entailments can be added to the original ontology. |
| Open Source Code | Yes | The test ontologies and our system are available online.1 [footnote 1: https://www.uni-ulm.de/en/in/ki/software/orar] |
| Open Datasets | Yes | For the popular benchmarks LUBM (Guo, Pan, and Heflin 2005) and UOBM (Ma et al. 2006), we use Ln and Un to denote the datasets for n universities respectively. |
| Dataset Splits | No | The paper evaluates the system on several real-world and benchmark ontologies, but there is no mention of training, validation, or test dataset splits. |
| Hardware Specification | Yes | All results were obtained using a compute server with two Intel Xeon E5-2660V3 processors and 512 GB RAM and a timeout of five hours. |
| Software Dependencies | Yes | We limit this comparison to the reasoners Konclude 0.6.2 (Steigmiller, Liebig, and Glimm 2014) and PAGOd A 2.0 (Zhou et al. 2015), which we found to perform best for our test ontologies |
| Experiment Setup | Yes | All results were obtained using a compute server with two Intel Xeon E5-2660V3 processors and 512 GB RAM and a timeout of five hours. |