Define OHDSI symposium workshop learning objectives for Data Catalog

[kzollove]

Catalog learning objectives

• How to identify relevant datasets from literature in your field; when to fall back on OHDSI GIS group for assistance
• get catalog entry in to spreadsheet (database?) (include variables in catalog??) | minimum required ...
• search catalog and add specific variables to ETL process
• share metadata with other research groups?

[jaygee-on-github]

• Determine what parts of a Dataset, the variables it measures and/or other schemas in a schema.org JSON-LD form our target metadata for the catalog entries. Especially when it comes to variables that measure exposures, we will learn how in schema.org JSON-LD we can represent their spatial and/or temporal coverage. In the process we hope to evaluate the use of Uber's H3 Hexagons.
• Determine how we will load the schema.org JSON.LD catalog entries into the GIS backbone. Here we will learn about the GIS backbone interfaces we can use to push metadata into one or more of the backbone tables.
• Decide on a standard input format for the schema.org JSON-LD catalog entries. Here we will study and learn which input formats users prefers when they describe exposure datasets.
• Provide step by step directions for building a mapping table that moves standard input into the schema.org JSON-LD. Here we will learn how to document and communicate mapping tables that are RDF documents themselves.
• Decide on a shape for the schema.org JSON-LD knowledge graph that captures all the catalog entries. Here we will learn how to develop ways to frame our knowledge graphs for fun and ease of use.
• Provide step by step instructions for navigating and/or querying this knowledge graph. Here maybe not in time for the global symposium we will learn how to make our knowledge graph queryable using an LLM. Note that LLMs are beginning to accept knowledge graphs as input.
• Capture the entire workflow in a YouTube. Here we will test maybe two approaches and learn from users which one is more successful. In one approach we might voice over a PowerPoint. In another approach we might narrate the execution of a Jupyter notebook.

[AEW0330]

Jay, this is great.
It would be great to pull out the things that would be listed in communications about the workshop that clarify what participants will learn. Those might take the form of items that complete the sentence: "After the workshop you will know how to..."

[tibbben]

Jay, this is great. It would be great to pull out the things that would be listed in communications about the workshop that clarify what participants will learn. Those might take the form of items that complete the sentence: "After the workshop you will know how to..."

• catalog learning objectives
  • get catalog entry in to spreadsheet (database?) (include variables in catalog??) | minimum required ...
  • search catalog and add specific variables to ETL process
  • share metadata with other research groups?