02 Integration and Semantic Mapping

Work in progress - current draft from working group - “CDIF Data Integration” - integration (Google Doc)

Group 02 Google Drive folder

 

Outline of this topic

  • Data integration has several aspects:

    • Combination of data at a structural level

    • Equating similar semantics related to the data

  • The roles played by concepts are the point of contact

  • Structural manipulation of data can be automated to a great extent. Mapping of semantics – even when informed by knowledge of the roles played by the relevant concepts – much less so.

  • What is the metadata needed to support these functions?

  • What is possible or desirable in terms of automation and support for non-automated activities?

  • What standards should be used? How does OWL fit into the picture?

  • Status:

    • Existing draft on structural aspects,

    • Initial exploration of SSSOM for semantic mapping

Suggested deliverables

  • Definition of the metadata needed for describing data or resopurces being exposed for cross-domain use

    • Structural metadata

    • Controlled vocabularies/semantics used

  • Structural metadata should be based on what exists in the originator’s systems

    • It should be encoded/modelled in a standard way

    • We are not describing the process of integration

  • Semantic mappings should be exposed as a resource to support integration for standard CVs/semantic schemes

    • Standard expression(s)/model(s) – e.g., SKOS, OWL and SSSOM

    • Standard encoding(s) – e.g., JSON-LD