=Paper=
{{Paper
|id=Vol-3939/abstract11
|storemode=property
|title=ON TO A Better Path to Choose Your Best Ontologies
|pdfUrl=https://ceur-ws.org/Vol-3939/abstract11.pdf
|volume=Vol-3939
|authors=Asiyah Yu Lin,John Graybeal,Anna Maria Masci,Juliane Schneider,Ruth Duerr,Eric G. Stephan,Hande Kuçük McGinty
|dblpUrl=https://dblp.org/rec/conf/icbo/LinGMSDSK24
}}
==ON TO A Better Path to Choose Your Best Ontologies==
https://ceur-ws.org/Vol-3939/abstract11.pdf
ON TO A Better Path to Choose Your Best Ontologies -
Abstract
Asiyah Yu Lin1, ∗, John Graybeal2, ∗, Anna Maria Masci3, Juliane Schneider4, Ruth
Duerr5, Eric G. Stephan4, Hande Kũçük McGinty6, ∗
1 Axle Research and Technology, Rockville, Maryland, U.S.
2 Graybeal.SKI Consulting, Stanford, California, U.S.
3 University of Texas MD Anderson Cancer Center , Huston, Texas, U.S.
4 Pacific Northwest National Laboratory,Richland, Washington, U.S.
5 Ronin Institute for Independent Scholarship, Montclair, New Jersey, U.S.
6 Kansas State University, Manhattan, Kansas, U.S.
Abstract
The development of biomedical ontologies has been rising exponentially in the last 10 years. In 2010,
181 ontologies were listed in BioPortal. As of June 2024, this number has increased to 1124, and many
other open ontology registration systems exist. When an ontology engineer starts developing an
ontology related to clinical studies, or anyone who wants to reuse similar ontologies, he/she may
need to navigate multiple similar ontologies, such as Clinical Study Ontology, Clinical Trial Ontology,
Clinical Research Ontology, SNOMED CT or NCIT, and more. A search of the term “clinical trial”
shows that 35 ontologies have the term labeled “clinical trial”.
This presents a challenge for data engineers and ontologists: how does one select the best fit-for-
purpose ontologies that work well for specific use cases? Often the selection process is unstructured
and highly biased, influenced by the domain experts involved, the background of the development
team, and the team’s theoretical approach to ontology adoption. While pursuing FAIR Principles for
ontology development, the reuse and interoperability of ontologies have been a serious concern
among ontologies, decision makers, and funders.
Beginning with work in the Research Data Alliance (RDA) Vocabulary and Semantic Services Interest
Group (VSSIG) over several years, our group has been gathering published and lived evaluation
techniques for choosing terms and ontologies. Here we share our observations of structured guidance
and specific criteria to choose the right ontologies that fit various semantic purposes. We distinguish
between criteria that can be evaluated with existing tools and services, criteria that can be evaluated
manually or subjectively, and criteria that could be evaluated with as-yet-unimplemented techniques.
Finally, we invite collaboration in ongoing development of the evolving materials through the RDA
VSSIG group, its slack channel, and its collaborative development in our Google Drive folder.
Keywords
Ontology, Vocabulary, Terms, Best Practices, Semantics, FAIR, Reusability1
15th International Conference on Biomedical Ontologies 2024, July 17-19, 2024, Enschede, The Netherlands
∗ Corresponding authors.
asiyah.lin@ohdsi.org(A.Y. Lin); jbgraybeal@sonic.net (J. Graybeal); amasci@mdanderson.org (A.M. Masci);
juliane.schneider@pnnl.gov (J. Schneider); ruth.duerr3@gmail.com (R. Duerr); eric.stephan@pnnl.gov (E.G.
Stephan); hande@ksu.edu (H.K. McGinty)
0000-0003-2620-0345 (A.Y. Lin); 0000-0001-6875-5360 (J. Graybeal); 0000-0003-1940-6740 (A.M. Masci); 0000-
0002-7664-3331 (J. Schneider); 0000-0003-4808-4736 (R. Duerr); 0000-0002-8155-6806 (E.G. Stephan); 0000-0002-
9025-5538 (H.K. McGinty)
© 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
ceur-ws.org
Workshop ISSN 1613-0073
Proceedings