Many research papers and tools do exist that automatically generate specific domain Ontology from text. and there exist tools to validate the OWL syntax and semantic relationships. However, do we have any tool that check the quality of the generated specific domain ontology?
Other than defining a a golden standard, I do not think this has been attempted yet.
You might start with the corresponding chapter of the Ontology Handbook (2nd edition):
http://userpages.uni-koblenz.de/~staab/Research/Publications/2007/OLBook2DellschaftStaab.pdf
http://userpages.uni-koblenz.de/~staab/Research/Publications/2007/OLBook2DellschaftStaab.pdf
There is a rather large body of work which looks at quality of ontologies, particularly in the biomedical domain but not exclusively.
For a tool you can just use there is the OntoOops online system: slides and link to online tool in slides: http://www.eudat.eu/system/files/MariaPovedaVillalon.pdf It looks for common issues with ontologies and flags them giving you an idea of quality.
Jeremy Rogers' early work on medical ontology quality: http://www.ncbi.nlm.nih.gov/pubmed/16685334
on concepulization quality:
http://www.sciencedirect.com/science/article/pii/S0306437908000689
and on activity levels as a metric:
http://www.ncbi.nlm.nih.gov/pubmed/22554701
and many more...
You can see the quality dimensions of ontology in this paper: https://www.researchgate.net/publication/235917545_The_Matrix_of_Quality_Dimensions_of_Knowledge_Management_Knowledge_Management_Assessment_Models_Review.
Article The Matrix of Quality Dimensions of Knowledge Management: Kn...
It is easier to address domain specific ontologies however you have to define what quality matters you are after?
Dear All,
Many thanks for your answers and suggestions. It is really helpful.
kind regards
I think this article may help you "Automatic Ontology Generation: State of the Art"
We gave a tutorial at CAISE 2010 about ontology quality (not very recent, I know, but most of the theoretical work on Ontology Quality was done in the 2000s). Here is the link to the presentation:
http://www.slideshare.net/palmonari/information-quality-in-the-web-era
You can find many relevant contributions to the topic at the end of the presentation, in the reference section.
There exists one tool to check ontology pitfalls devloped by UPM:
http://oeg-lia3.dia.fi.upm.es/oops/index-content.jsp
You can find on the Web several papers and additional material that describes OOPS and background theory. Here are some examples:
http://oa.upm.es/21895/1/The_Current_Landscape_of_Pitfalls_in_Ontologies.pdf
http://oa.upm.es/10195/1/OOPS_technical_report_v0.2.pdf
I think that a Journal paper about OOPS will be published soon - if not published yet.
Finally, to broaden up a little bit the topic, let me mention a couple of papers I co-authored that address the topic of evaluating the quality of information in the wild. In these papers we also discuss some theoretical aspects of quality in Knowledge Bases, including ontologies.
Carlo Batini, Matteo Palmonari, Gianluigi Viscusi: The Many Faces of Information and their Impact on Information Quality. ICIQ 2012: 212-228
You can find the paper in the ICIQ proceedings:
http://mitiq.mit.edu/ICIQ/2012/2012%20ICIQ%20CDproceedings%20final.pdf
Carlo Batini, Matteo Palmonari, Gianluigi Viscusi: Opening the Closed World: A Survey of Information Quality Research in the Wild. In Floridi, L. and Illary, P (Eds.): The Philosophy of Information Quality. pp. 43-73. Springer. 2014.
http://link.springer.com/chapter/10.1007/978-3-319-07121-3_4#
(BTW: this book contains many other interesting contributions about information quality).
Best,
Matteo
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Graphs