When DL-Lite met OWL

DL-Lite is a family of Description Logics (DLs) whose aim is to capture some of the most popular conceptual modeling formalisms, such as Entity-Relationship model [4] and UML class diagrams1, while preserving the tractability of the most important reasoning tasks, such as ontology satisfiability and query answering of arbitrary (union of) conjunctive queries (ground and not ground). More specifically, reasoning over ontologies of the DL-Lite family, is LOGSPACE in data complexity, and can be entirely delegated to a standard DBMS technology. We refer here to DL-LiteA, the DL of the DL-Lite family that is closest to OWL 2.0. Note that DL-LiteA is in fact a fragment of OWL 2. As such, it is possible to restrict the syntax of OWL so that it captures DL-LiteA. Because of the lack of the space, we will not discuss further this topic. We provide an expressive query language for DL-LiteA ontologies, named SparSQL , whose syntax is inspired by both SQL and SPARQL3, one of the most popular W3C proposals as standard query language for OWL. Actually, SparSQL is an epistemic query language that implements the language EQL-Lite(UCQ) presented in [2]. Thus, query answering is in LOGSPACE and can be reduced to evaluate first-order logic queries over the ABox. Moreover, we stress the expressive power of DL-LiteA by presenting a series of features, that are beyond OWL and can be added to the language. Specifically, these features allow one to express data properties of object properties, denial and identification constraints, and they include a new ontology component, called the ECBox, which is a set of general form of constraints called EQL constraints, based on epistemic logics. Note that this is not the first attempt to add constraints to OWL. However, previous related work, e.g. [5], are technically and semantically incomparable with the approach presented in this paper. Notably, all the contributions discussed in this work are currently implemented within the MASTRO system, a tool for ontology representation and reasoning that has DL-LiteA as proprietary core language. The main feature of MASTRO is to reduce all reasoning tasks, such as consistency checking and query answering to the evaluation of standard SQL queries over a DBMS. Note that one of the major benefits of using