GraphDB

Metadata Studio keeps its data and state in GraphDB, which has the following advantages:

  • out-of-the-box integration of the data with the Ontotext Platform GraphQL interface;
  • closeness to reference data, which enables unified access to the data during annotation as well as analyzing the impact of reference data changes over the gold standard corpus;
  • making use of the GraphDB Text Mining plugin and Connectors;
  • easy administration thanks to the straightforward setup, backup, and restore.

The following diagram presents how the data is distributed across the GraphDB repositories:

https://lucid.app/publicSegments/view/0c263f7e-be47-4c94-9187-8618cf544e93/image.jpeg

where the Reference data can also be accessed via federation from an external SPARQL endpoint.