CINERGI / ontology_cleanup

Using OWL API to clean up ontologies
0 stars 0 forks source link

9/25 processing: to be added to the ontology #2

Open izaslavsky opened 8 years ago

izaslavsky commented 8 years ago
adamschachne commented 8 years ago

What is Rock Core a synonym of?

Granite among many other rocks should already be traversing the Material > Rock material search path. Granite remains as is, since it is already a subclass of Rock material.

Geothermal cinergiPreferredLabel is Geothermal Feature

izaslavsky commented 8 years ago

I think "Rock Core" would belong to "Activity - Observation". it really refers to a kind of sample. Steve, what do you think?

On Tue, Sep 29, 2015 at 3:18 PM, adamschachne notifications@github.com wrote:

What is Rock Core a synonym of?

Granite among many other rocks should already be traversing the Material > Rock material search path. Granite remains as is, since it is already a subclass of Rock material.

Geothermal cinergiPreferredLabel is Geothermal Feature

— Reply to this email directly or view it on GitHub https://github.com/CINERGI/ontology_cleanup/issues/2#issuecomment-144207127 .

smrgeoinfo commented 8 years ago

a 'rock core' is a kind of specimen. From the point of view of the cinergi_facets.owl ontology, the most coherent treatement I can think of is to add 'physical sample' as a kind of Resource_Type, and put specimen, rock sample, rock core, thin section, and things like that in that class

izaslavsky commented 8 years ago

Hmmm... I think in our case we are not really populating resource types through the keyword mechanism (Alice and Adam, correct me if you find that you are making such assignments). "Resource type" assignment would mostly come from analysis of other metadata elements (eg distribution, mime type,

On Wed, Sep 30, 2015 at 12:25 PM, Stephen Richard notifications@github.com wrote:

a 'rock core' is a kind of specimen. From the point of view of the cinergi_facets.owl ontology, the most coherent treatement I can think of is to add 'physical sample' as a kind of Resource_Type, and put specimen, rock sample, rock core, thin section, and things like that in that class

— Reply to this email directly or view it on GitHub https://github.com/CINERGI/ontology_cleanup/issues/2#issuecomment-144514021 .

smrgeoinfo commented 8 years ago

ontology items mapped to the 'observation' facet: Category Observation Complex Observation Count Observation Discrete Coverage Observation Geometry Observation Measurement Temporal Observation Truth Observation rock core doesn't fit too well. rock core could be thought of as a sampling feature, but doesn't fit into our feature facets. Maybe add 'observation target' under observation for 'rock core' and related samplingFeatures?

izaslavsky commented 8 years ago

I like observation target! Or shall we call them "features"? Features of interest are under "features", but these would be different "features", under observations. To what extent do we follow OGC lingo here. Targets seems to work around this problem.

On Wed, Sep 30, 2015 at 12:47 PM, Stephen Richard notifications@github.com wrote:

ontology items mapped to the 'observation' facet: Category Observation Complex Observation Count Observation Discrete Coverage Observation Geometry Observation Measurement Temporal Observation Truth Observation rock core doesn't fit too well. rock core could be thought of as a sampling feature, but doesn't fit into our feature facets. Maybe add 'observation target' under observation for 'rock core' and related samplingFeatures?

— Reply to this email directly or view it on GitHub https://github.com/CINERGI/ontology_cleanup/issues/2#issuecomment-144519319 .

smrgeoinfo commented 8 years ago

I was thinking that 'observation target' would resonate better with non O&M aligned people, as well as make more sense under the 'observation' facet than sampling feature. I'm good with it if you are!

valentinedwv commented 8 years ago

Sample Media, or Sample Type?

izaslavsky commented 8 years ago

There is sample processing ontology http://bioportal.bioontology.org/ontologies/SEP

Perhaps there is an ODM2 vocabulary we can use here? since we need specimen, time series, sections, etc. - all of them.

On Wed, Sep 30, 2015 at 1:38 PM, David Valentine notifications@github.com wrote:

Sample Media, or Sample Type?

— Reply to this email directly or view it on GitHub https://github.com/CINERGI/ontology_cleanup/issues/2#issuecomment-144536264 .

valentinedwv commented 8 years ago

Don't see how the sample processing ontology would work. Focused on gels.

Can use a few ODM2 ones: http://vocabulary.odm2.org/specimentype/ http://vocabulary.odm2.org/resulttype/ http://vocabulary.odm2.org/medium/

izaslavsky commented 8 years ago

Indeed. Adam, could you add terms from the first one ( http://vocabulary.odm2.org/specimentype/) under Observations?

On Wed, Sep 30, 2015 at 2:53 PM, David Valentine notifications@github.com wrote:

Don't see how the sample processing ontology would work. Focused on gels.

Can use a few ODM2 ones: http://vocabulary.odm2.org/specimentype/ http://vocabulary.odm2.org/resulttype/ http://vocabulary.odm2.org/medium/

— Reply to this email directly or view it on GitHub https://github.com/CINERGI/ontology_cleanup/issues/2#issuecomment-144552732 .

izaslavsky commented 8 years ago

Then 'rock core" will be a synonym for "core" in this list list. Use "names" in this list as labels.

On Wed, Sep 30, 2015 at 3:24 PM, Ilya Zaslavsky ilya.zaslavsky@gmail.com wrote:

Indeed. Adam, could you add terms from the first one ( http://vocabulary.odm2.org/specimentype/) under Observations?

On Wed, Sep 30, 2015 at 2:53 PM, David Valentine <notifications@github.com

wrote:

Don't see how the sample processing ontology would work. Focused on gels.

Can use a few ODM2 ones: http://vocabulary.odm2.org/specimentype/ http://vocabulary.odm2.org/resulttype/ http://vocabulary.odm2.org/medium/

— Reply to this email directly or view it on GitHub https://github.com/CINERGI/ontology_cleanup/issues/2#issuecomment-144552732 .

adamschachne commented 8 years ago

Is there an ontology for these terms, or should I create new classes?

izaslavsky commented 8 years ago

just create new classes. I don't think there is an ontology. There might be a SKOS.