A Resource Description Framework (RDF) provides a framework for specifying–in digital library terminology, metadata–that makes information machine-readable. It is based on a graph with subjects, predicates, and objects. The basic statement has a subject in the form of a Uniform Resource Identifier (URI), and an object that can be either a URI or a literal value. Together these are known as a triple.
RDF enables information to be described in a machine-readable way that software can interpret. It is the basis of technologies such as SPARQL, RDF Schema (RDFS), SKOS, and Web Ontology Languages (OWL). It describes resources in a way that is unambiguous to the software working with it. It does so by describing relationships between subjects, predicates, and objects. Each subject, predicate, and object is expressed by a triple (subject, attribute, value) that can be mapped to a data structure like XML.
For example, a triple may say that a book has a title and three authors. An XML file representing this description might look something like Figure 6.5. However, the XML can be converted into an RDF graph that can then be used by software to perform queries over it. It is this ability to use RDF to describe information that enables the Semantic Web and Linked Data. It is the reason that RDF is often used as a metadata framework for the Web.
An RDF statement (also known as a triple) is a labeled, directed graph. An arc between two nodes in the graph represents a relationship between those two things. The nodes represent subjects or objects, and the arrows represent properties (sometimes called predicates). Each object can be identified by one of three types of identifiers: a Uniform Resource Identifier (URI), a literal value, or a blank node. The URIs used in RDF can be resolvable with a standard web protocol, and they may contain characters from any of the world’s languages.
The RDF model is so flexible and simple that it has led to its use in applications unrelated to the Semantic Web activities of the W3C. It makes it easy to combine information from different sources with a single syntax like Turtle or N-Triples. RDF also supports a variety of ways to store and retrieve information in a database, including popular triplestores such as Ontotext’s GraphDB.
In RDF, relationships between data are represented as triples – subjects, predicates, and objects. A subject is a URI, a predicate is an RDF property and the object is a resource. The triples can be combined to form graphs, which provide the basis for a basic level of inference. The semantics of the relationships are provided by the RDF Schema. There are also more sophisticated levels of inference available, using richer languages such as OWL [OWL2-OVERVIEW] and inference rules.
One of the more important and popular W3C standards, RDF is a data format for machine-readable, contextual, and semantically rich information. It is used as the foundation for Linked Open Data and provides the building blocks for a web of data that can be interconnected across all platforms, languages, and applications. RDF has features that make it easy to merge datasets even if their underlying schemas differ and supports the evolution of schemas over time without requiring all the data consumers to change (another powerful feature for today’s fast-changing world).
Click here to read the original article published by Semantic Web Server.
Please give your feedback on this article or share a similar story for publishing by clicking here.
