SHACL validation support in Whyis

[mdeagen]

Should we add a bounty for integrating a SHACL validator in Whyis?

A data curation pipeline could pass an incoming dataset through various SHACL shape graphs and display validation reports (warnings, etc.), to assist the curator in data cleaning or deciding whether a virtual column is needed (in the case of semantic data dictionary curation).

A new version of pySHACL was released a couple weeks ago https://github.com/RDFLib/pySHACL

[jpmccu]

I like this. The validator can mark the relevant nanopublications and link to the root entities that are examined. This would be a good first attempt at creating quality reports for the graph and do the initial modeling for that and documenting "issues" in the graph. Another one would be the Ontology hygiene checks (they're all sparql queries).

…

On Fri, Sep 18, 2020 at 10:51 AM mdeagen ***@***.***> wrote: Should we add a bounty for integrating a SHACL validator in Whyis? A data curation pipeline could pass an incoming dataset through various SHACL shape graphs and display validation reports (warnings, etc.), to assist the curator in data cleaning or deciding whether a virtual column is needed (in the case of semantic data dictionary curation). A new version of pySHACL was released a couple weeks ago https://github.com/RDFLib/pySHACL — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#237>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAETCEJPAEQIIJAAOHNVU5TSGNXYDANCNFSM4RSDGZSQ> .

-- Jamie McCusker (they/them/theirs) Director, Data Operations Tetherless World Constellation Rensselaer Polytechnic Institute mccusj2@rpi.edu <mccusj@cs.rpi.edu> http://tw.rpi.edu