TechBytes 4 May 2017

Introduction to RDF

Introducing (RDF) that is made of 3-part subject, predicate, object statements called triples. The big shift for libraries is going from managing MARC records to managing RDF triples

Linked Data in libraries is constructed using Resource Description Framework (RDF) graphs, a type of directed graphs, where the nodes are made up of either IRI (international resource indicator i.e. URLs or URIs), blank nodes (a type of identifier placeholder in a RDF graph), or literal values.

Originating with the World Wide Web Consortium (W3C) specifications, RDF graphs are built using three element statements called triples that model relationships between resources, IRIs, and descriptive information.

In a RDF triple, the first element is called a subject and represents a resource with the second element, the predicate, describing an aspect that connects to the third element, an object made up a value. One or more triples make up a RDF graph.


Subjects must be either an IRI or a blank node and represents the resource or entity



A relationship between the subject and object is a predicate, also called a property. A predicate can only be an IRI.


  • <>
  • <>
  • <>


Object must be either an IRI, blank node, or literal value.


Example RDF Graph in Turtle Serialization

@prefix bf: <> .
@prefix owl: <> .
@prefix rdf: <> .
@prefix rdfs: <> .
@prefix schema: <> .
@prefix xml: <> .
@prefix xsd: <> .

<> a schema:CollegeOrUniversity ;
    rdfs:label "University of Denver"@en ;
    schema:department [ a bf:Organization ;
            rdfs:label "Library and Information Science"@en ;
            bf:subject <> ;
            owl:sameAs <> ] .