Proceedings Abstracts of the Twenty-Fifth International Joint Conference on Artificial Intelligence

Object-Relational Queries over CFDInc Knowledge Bases: OBDA for the SQL-Literate / 1258
Jason St. Jacques, David Toman, Grant Weddell

We consider how SQL-like query languages over object-relational schemata canbe preserved in the setting of ontology based data access (OBDA), thus leveraging wide familiarity with relational technology. This is enabled by the adoption of the logic CFDnc-forall, a member of the CFD family of description logics (DLs). Of particular note is that this logic can fully simulate DLlite-F, a member ofthe DL-Lite family commonly used in the OBDA setting. Our main results present efficient algorithms that allow computation ofcertain answers with respect to CFDnc-forall nowledge bases, facilitating direct access to a pre-existing row-basedrelational encoding of the data without any need for mappings to triple-based representations.