Search Mailing List Archives
[bioontology-support] [BioPortal] Feedback from Clement Jonquet
Paul R Alexander
palexander at stanford.edu
Mon Jun 23 11:37:20 PDT 2014
Thanks for reporting this. By default, we only return one class per match per ontology when using longest-only. We haven’t seen much of a use case for doing otherwise because we felt that ontologies typically wouldn’t have overlapping concepts. However, I’ve created an issue for us to investigate doing something better than choosing the first class from the results, which I believe is what we’re doing now.
On Jun 23, 2014, at 4:30 AM, support at bioontology.org wrote:
> Name: Clement Jonquet
> Email: jonquet at lirmm.fr
> Location: http://bioportal.bioontology.org/annotator
> While testing the new longets only parameter of the Annotator we found out something strange:
> Try to annotate the text "cell" with ontology MESH.
> longestOnly=false returns 2 annotations
> longestOnly=true returns 1 annotation .
> Cell (T025)...
> It seems there is something weird (maybe because the annotation remove is a semantic type...
> Can you reproduce and tell us more.
> Clement and Soumia
> bioontology-support mailing list
> bioontology-support at lists.stanford.edu
More information about the bioontology-support