Data Integration Using RDF #MRDinteg

The MRDinteg Group was one of five Working Groups formed at the JISCMRD Phase Two Launch Workshop.

The MRDinteg Group set out to discuss shared experiences of data integration using RDF.

As well agreeing to share ideas and experiences, the Group agreed that it would be a good thing to organise a Hack Day centred around tools for building and manipulating triplestores, based on the projects’/group’s experiences. This would likely come under DevCSI umbrella.

Discussions ranged over the approach being taken (e.g. the idea of a triple store as an overlay rather than replacement for data sources; the need to maintain the appropriate balance between imposing common ontological standards and retaining the freedom to specialise away from what common standards do) and the areas where experiences could constructively be shared.

The latter included:

• The process of mapping data sources to ontologies: What tools have we found available and useful for doing data mapping?
• What king of editorial controls to exercise when integrating data, and where to apply them?
• How to go about exploring and navigating the resulting integrated data (level of tools and what should the navigation look like?

The group identified the following specific areas for collaboration and action:

• Investigate workflows for getting data in to triple stores.
• Capture provenance data quality information.
• Share / develop models for representing provenance.
• Explore multi-lingual support for queries and data.