How Long Will It Take? Accurate Prediction of Ontology Reasoning Performance

Companion Page

Yong-Bin Kang, Jeff Z. Pan, Shonali Krishnaswamy, Wudhichart Sawangphol and Yuan-Fang Li


This is the companion web page for our paper titled "How Long Will It Take? Accurate Prediction of Ontology Reasoning Performance" accepted at AAAI 2014. If you have any questions regarding the research work or software, please feel free to contact me.

The Ontologies

On this page we provide the dataset we use in the paper. The dataset contains 451 ontologies we collected from the public domain. They are organized in a compressed directory and can be downloaded here (64.1 MB compressed, 1.28 GB uncompressed). More details of these ontologies and how we use them can be found in the paper.

The reasoners

The performance of 6 reasoners are evaluated in our paper: FaCT++ (version 1.5.3), HermiT (version 1.3.6), JFact (version 0.9), MORe (version 0.1.6), Pellet (version 2.2.0) and TrOWL (0.8). For the task of classification, the user time of each reasoner is recorded.

Metric values

The following csv files record the metric values and runtime for all the ontologies and for all the 6 reasoners.

Note that a 50,000-second cut-off time is applied on all reasoners.

Prediction models

The prediction models for the 6 reasoners trained on the above dataset can be found here. The Random Forests models are developed using the R packages randomForest and caret.