OHDSI / GIS

https://ohdsi.github.io/GIS
Apache License 2.0
10 stars 9 forks source link

Create a proof of concept knowledge graph for a catalog that includes both SDoHs and climate data #326

Open kzollove opened 9 months ago

kzollove commented 9 months ago

Create a proof of concept knowledge graph for a catalog that includes both SDoHs and climate data. At least one SDoH entry will have an API. At least one climate entry will be for raster data. Spatial coverage of maybe half the catalog entries in the proof of concept will be for the Global South

jaygee-on-github commented 8 months ago

The ESIP (Earth Sciences Information Partners) Federation has developed some "experimental recommendations" for mostly variable level metadata which we are reviewing.

One recommendation goes to documenting the range of a variable when its value comes from a controlled vocabulary.

fils commented 6 months ago

We have been working with the ERDDAP developer community in NOAA and might be able to find some related data. It would definitely be raster data. Again, more ocean than climate though. So might not fit your use case.

I do have a fair bit of ocean data from the global south I could provide that does have spatial data, though much of it is unstructured or tabular not raster, so might not be of interest. We have data from the Pacific Data Hub (PDH) and Research Data Australia (RDA). The PDH has some social-economic data too they have been interested in connecting with our more ocean centric graph resources.

@jaygee-on-github , that PDH data is something Arofan is working with so if you ask him about he can tell you real details about it. I don't know if it is SDoH related or not.

jaygee-on-github commented 6 months ago

@fils can you connect with @tibbben maybe now or maybe later (after Friday) -- he will have some tabular climate data -- so do "we" where we is APHRC. We are using it with deaths from malaria. See #289. Let's target some SDOH data on Friday with everyone else. One thought here is we build a knowledge graph across the dataset and variable levels in a proof of concept using #311. In the meantime don't worry about raster. Just use a dataset that you have used before even if it is from the ocean. Smile.

Before Friday, however, check out #323. Here @p-talapova inventories some of our dataset use cases and indicates how we might map them to "our" vocabularies downstream -- either an OHDSI standard one or one that her team has developed and/or is proposing. That way you will know our world and be able to help us think through what datasets we want to include in this proof of concept.

Thanks.

@kzollove and @martyalvarez, the presentation/conversation on Friday could take more time than we originally thought.