Small Is Beautiful: Computing Minimal Equivalent EL Concepts
Authors: Nadeschda Nikitina, Patrick Koopmann
AAAI 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In this paper, we present an algorithm and a tool for computing minimal, equivalent EL concepts wrt. a given ontology. Our tool can provide valuable support in manual development of ontologies and improve the quality of ontologies automatically generated by processes such as uniform interpolation, ontology learning, rewriting ontologies into simpler DLs, abduction and knowledge revision. Deciding whether there exist equivalent EL concepts of size less than k is known to be an NP-complete problem. We propose a minimisation algorithm that achieves reasonable computational performance also for larger ontologies and complex concepts. We evaluate our tool on several bio-medical ontologies with promising results. |
| Researcher Affiliation | Academia | Nadeschda Nikitina University of Oxford nadeschda.nikitina@cs.ox.ac.uk Patrick Koopmann University of Dresden patrick.koopmann@tu-dresden.de |
| Pseudocode | Yes | Algorithm 1: MINIMISE function computing a term representing a minimal equivalent concept. Algorithm 2: COMPUTERULES function computing rules of type DR2 for RE,k,C,O |
| Open Source Code | No | The paper mentions 'Our tool' but does not provide any link or explicit statement about its source code being publicly available. |
| Open Datasets | Yes | We evaluate our method on concepts from Snomed Clinical Terms (Snomed CT) (Stearns et al. 2001), National Cancer Institute Thesaurus (NCIT) (Sioutos et al. 2007), Galen (Rector et al. 1994), Fission Yeast Phenotype Ontology (FYPO) (Harris et al. 2013) and Genomic Clinical Decision Support Ontology (Genomic CDS) (Samwald 2013). |
| Dataset Splits | No | The paper describes selecting concepts for evaluation and minimization but does not specify training, validation, or test dataset splits in the conventional sense for model training. |
| Hardware Specification | No | The paper mentions the 'ELK reasoner' and 'OWL API' but provides no specific hardware details (e.g., CPU, GPU models, memory) used for running the experiments. |
| Software Dependencies | No | The paper mentions 'ELK reasoner (Kazakov, Kr otzsch, and Simanˇc ık 2014)' and 'OWL API (Horridge and Bechhofer 2011)' but does not provide specific version numbers for these software dependencies. |
| Experiment Setup | Yes | For each ontology O, we selected 100 EL concepts occurring in the axioms of O with the size at least 2. ... We set a timeout of 5 minutes for all ontologies except Snomed CT, for which we increase the timeout to 30 minutes due to longer reasoner response times. |