RDF is a W3C standard for describing information about a resource on the Web. RDF is meant to be read and understood by computers.
While it does not directly contain support for generating external or embedded RDF, DITA Open Toolkit does have some functionality that can be used to create RDF.
The Dublin Core is a standard for metadata that is used to describe online information. The XHTML output produced by DITA Open Toolkit contains Dublin Core metatags generated from the various elements contained within the prolog, title, amd short description elements in DITA source files. Further processing of the XHTML output can create RDF "triples" using these meta tags. (Functionality for that processing is not contained in the Toolkit.)
An RDF triple contains three pieces of information, a subject, a property type, and a value for the property.
<meta name="DC.Title" content="About metadata"/>
In this example the triple says the web page name is "About metadata".
The Thesaurus (aka Taxonomy) plug-in can be installed with DITA Open Toolkit to provide a DITA specialization that can be used to identify and process content based on what the information is about by generating SKOS output. See About DITA Open Toolkit plug-ins for more information about this plug-in.