search

skybristol / geokb

Data processing workflows for initializing and building the Geoscience Knowledgebase
The Unlicense
3 stars 3 forks source link

Align GeoKB geologic time classification with GSO #28

Closed skybristol closed 10 months ago

skybristol commented 10 months ago

We should be in basic alignment with how the GSO organized geologic time since we used the same underlying source, but the OWL representation of relationships is different and we need to record the GSO's gst namespace identifiers.

skybristol commented 10 months ago

One of the main things to examine here is the shift from SKOS to OWL between the GeoSciML version of the ICS chart I used previously and the OWL approach in the GSO. I need to look at the whole broader/narrower approach I've used here and with some aspects of the USGS Thesaurus vs. the notion of potentially more specific properties describing a hierarchy that will likely serve us better.

We also need to examine whether there are any other aspects of descriptive information on geo-time we want to incorporate into the GeoKB. For instance, we may want to at least indicate when there has been a change in definition over time through the ICS even though we may not record the details in our own representation.

A third area of focus will be on improving the linkages out from our simplistic representation of named geologic time units to other sources, including Wikidata.

A fourth area of focus will be on any additional attribution or describing specific query language needed to pull geologic time units for a specific purpose such as a code list for the sample submittal module of the upgraded Laboratory Information Management System being put in place for the Analytical Geochemistry Lab.

skybristol commented 10 months ago

Mike Zientek in our group brought up the fact that published USGS information products are instructed to conform to the consensus reached by the USGS Geologic Names Committee and released through this fact sheet in 2018. As we rework the classification scheme, we'll examine any deviation from the ICS Chart and potentially introduce additional classification to enable ready query of the USGS "Chart of the Divisions of Geologic Time."

skybristol commented 10 months ago

For reference, this item represents that USGS-GNC source, pulled previously in the Pubs Warehouse harvest:

https://geokb.wikibase.cloud/wiki/Item:Q57435

One approach may be to simply record the reference on the subClassOf claims for those geologic time units that align, but we'll have to see where the deviations are.

skybristol commented 10 months ago

The results of this work and useful queries in the GeoKB are documented on the upper classification item for geologic time interval.