Data Integration

Created on Feb. 18, 2013, 1:39 p.m. by Hevok & updated by Hevok on May 2, 2013, 4:42 p.m.

Assuming one need a Database (e.g. with bibliography data) into a RDF knowledge base (e.g. tables on Articles, Authors and Journals). Such highly structured data as it is in Relation Databases (i.e. tables) can rather easily by transformed into a RDF-Graph. There are general rules how an RDF-Graph can be created from the bases of a simple Relational Database.

RDF is rather useful especially for the purpose of Data Integration.

One just exports the Database into a Set of Relations.

If one has two databases with a different Schema definition, i.e. different structure, then it is pretty difficult to map the data inside these tables together. So Schema mapping is a rather difficult problem. In RDF Schema mapping becomes much more easier.

To map data in the traditional sense would be rather complicated, but in RDF this becomes very simple as one has to map the things that have the same URI. For example consider additional data. Thus one need to identify at least two identities. By identifying the same identities one can connect the graphs.

Data Integration is a pretty simple thing to be made in RDF, especially if compared to traditional relational data.

operational-data -integration.jpg

Tags: databases, unification, linking, data
Children: Information Integration, RDF Data Model

Update entry (Admin) | See changes

Comment on This Data Unit