Suppose I create a new context / data source for RDFs.
Then I use the APIs to feed in a RDF file. Had I gone through the TRIT, the uris for the predicate and entities are automatically created.
In this case, how do I generate ‘appropriate’ uris for new predicates/ entities if not going through the TRIT ? Do i generate any random guid to suffice?
Correct, any valid IRI according to the RDF spec will do. For more details see
https://www.w3.org/TR/rdf11-primer/
To properly visualize the relationships in DE/Analyze the system expects CamelCase in IRI fragment, e.g.,
http://rdf.thomsonreuters.com/relationship/IsSupplierOf
becomes
is supplier of
When using a JDBC connector or delimited import these are automatically created in the data source context based on the ID selected by the user. Context + ID form the subject IRI.
One caveat here is that if you use a subject IRI that already uniquely identifies another entity in the system, e.g.,
http://permid.org/1-4295905573
the system will automatically treat all entities using this subject IRI as one entity, even if they were loaded into different data sources.
