Creating scalable, dynamic applcations using Linked Data
An ontology, exposed in any open XML format (zThes, SKOS, OWL) provides the lightweight semantics for Linked Data. The issue we see is that for a Linked Data application to use the information in an ontology it needs to be able to query the term via an HTTP service - a "dereferencable URI".
Semantic Enhancement Server - a dynamic ontology server
If the ontology can be modeled in Ontology Manager then it can be served in the Semantic Enhancement Server. This optimized term index has two key advantages:
- The service is highly scalable. It has been stress tested to 200 term queries per second with sub-second response on a single CPU machine. The alternative is typically parsing large static XML files on demand, which has a high overhead.
- The service can provides dereferencable URI access and returns multiple output formats. All the core concept details, concept metadata and relationship information is returned, and transformed on the fly to the required output standard. This makes it easy for an application to get to the ontology – the work is done by the ontology service meaning the application does not have to worry about translating and mapping the xml returned.