Is Envo curation linked at all to Gazetteer?

[Public-Health-Bioinformatics]

I ask because in Gazetteer US states are linked as "national political entity", but I think a new term is needed for this: "subnational political entity". Right now the US states are in the same group as countries/nations.

<owl:NamedIndividual rdf:about="&obo;GAZ_00002518">
    <rdf:type rdf:resource="&obo;ENVO_00000009"/>
    <obo:RO_0001025 rdf:resource="&obo;GAZ_00003861"/>
    <obo:IAO_0000115 rdf:datatype="&xsd;string">A state located in the southwestern region of the United States. It borders New Mexico, Utah, Nevada, California, touches Colorado, and has a 626 km international border with the States of Sonora and Baja California in Mexico.</obo:IAO_0000115>
    <rdfs:label rdf:datatype="&xsd;string">State of Arizona</rdfs:label>
</owl:NamedIndividual>

[cmungall]

See #65 (sorry about the awful formatting from the google code import)

[cmungall]

Earlier text mining experiment here: https://github.com/EnvironmentOntology/envo/tree/master/experimental

Would love to work on this again..

[cmungall]

Note that I managed to get a property into wikidata for mapping WD URIs to ENVO classes: https://www.wikidata.org/wiki/Wikidata:Property_proposal/ENVO_ID

The idea is we can then load a CC-0 subset of ENVO into WD, we then get linkages to instances via WD instance-of edges - and we can map these to GAZ (or simply use WD directly in place of GAZ; should do a gap analysis, lots in GAZ that isn't in WD/WP...)

[Public-Health-Bioinformatics]

I'd welcome a greater integration or WD use as an OBOFoundry sanctioned resource esp. if GAZ isn't being supported in curation anymore. We're needing it for its global region, national, subnational and municipal terms.
About the textmining - seems like the word "state" isn't really used in GAZ nation labels, so maybe drop that rule due to its false positives? Also, I see now the EBI version of GAZETTEER is different - the OBOFoundry version annotation rdf:type ... is missing? So its the textmining relations that have been added on automatically in OBOFoundry?

[cmungall]

On 22 Jun 2017, at 14:37, Damion Dooley wrote:

I'd welcome a greater integration or WD use as an OBOFoundry sanctioned resource esp. if GAZ isn't being supported in curation anymore. We're needing it for its global region, national, subnational and municipal terms. About the textmining - seems like the word "state" isn't really used in GAZ nation labels, so maybe drop that rule due to its false positives?

There are probably a few of these.

Also, I see now the EBI version of GAZETTEER is different - the OBOFoundry version annotation rdf:type ... is missing? So its the textmining relations that have been added on automatically in OBOFoundry?

Ah, I had forgotten about this. I recall that very conservative mined relations were added, but this was a while ago.

-- You are receiving this because you commented. Reply to this email directly or view it on GitHub: https://github.com/EnvironmentOntology/envo/issues/514#issuecomment-310509786

[pbuttigieg]

The idea is we can then load a CC-0 subset of ENVO into WD, we then get linkages to instances via WD instance-of edges - and we can map these to GAZ (or simply use WD directly in place of GAZ; should do a gap analysis, lots in GAZ that isn't in WD/WP...)

This sounds very cool and has potential to get some degree of multilingualism for ENVO terms.

[Public-Health-Bioinformatics]

I guess this should be closed then as solution is in the adjustment of external ontologies.

[pbuttigieg]

Hi @ddooley I'm closing this, but do direct the GAZ curators to this discussion in the GAZ tracker.