SPARQL in N3: Translating SPARQL into a Semantic Rule Language to Facilitate Logic Reasoning on the Semantic Web

Rule-based reasoning on the Semantic Web (SW) is a logics-based paradigm for authoring expressive, exchangeable and interoperable rules. Such rules have the ability to natively access, operate on, and reason over online and interconnected data sources. While the full potential of rule-based reasoning has not yet been tapped, query-based data access is well developed and mature on the SW. The SPARQL Protocol and RDF Query Language (SPARQL) is a well-known and widely used standard for querying SW data. An option is thus to express rule-based reasoning using SPARQL queries to leverage its popularity and ease-of-use. Prior approaches to utilize SPARQL queries in this way are either limited in expressivity, operational semantics (bottom-up reasoning), or rely on a target rule language with a significant impedance mismatch (Datalog). We present an approach that instead translates SPARQL queries to the Notation3 (N3) SW rule language. We adhere to a one query–one rule principle, where each SPARQL query is translated into a single N3 rule, in order to improve the exchangeability of translated rules. We formally prove the correctness of our approach and evaluate the performance of our approach compared to the state of the art. We view our work as a first step in the cross-pollination of SW query and rule languages.