Querying the Semantic Web: SPARQL

Abstract

SPARQL – Simple Protocol And RDF Query Language – is the language, proposed by W3C, for querying RDF data published on the Web, both stored natively or viewed via middleware. SPARQL offers a syntactically SQL-like language for querying RDF graphs via pattern matching, as well as a simple communication protocol that can be used by clients for issuing SPARQL queries against endpoints. The first section provides the reader with a scientific and technical overview of the SPARQL query language. Basic concepts, such as the notions of triple and graph patterns are presented first. The section, then, shows how to write simple queries, and progressively introduces the reader to the full expressive power of SPARQL. The second section illustrates some examples of applications, progressing in a quasi-chronological order. It starts with “early days” applications, when RDF data were lacking and the Semantic Web practitioners applied semantic technologies to bibliographic and conference data. Next, it moves on to “first uptakes” in the area of bioscience, which can be considered as the earlier science adopting Semantic Web technologies. This section is concluded by the presentation of some large applications, showing SPARQL queries that nowadays can be issued against interlinked RDF repositories about music and about governmental data. The third section is dedicated to SPARQL implementations, in particular to those ones that are mostly used and widely deployed. It also discusses the standard compliance of the implementations, based upon the W3C test suite. Finally, the fourth section discusses some of the issues that characterize the current development of SPARQL. It presents foreseen extensions to the query language, in particular a proposal for remotely updating RDF graphs, four vocabularies for describing SPARQL endpoints, the behavior of SPARQL under different entailment regimes, three approaches for querying the entire Semantic Web with SPARQL, and three proposals for extending SPARQL to the management of streams of rapidly flowing information.