Data and Knowledge Group

― Knowledge Representation and Reasoning



REQUIEM (REsolution-based QUery rewrIting for Expressive Models) is a prototypical implementation of a query rewriting algorithm developed by Héctor Pérez-Urbina, Boris Motik, and Ian Horrocks.

Given an ELHI TBox and a conjunctive query, REQUIEM computes a datalog query that is a rewriting of the query w.r.t. the TBox. REQUIEM produces worst-case optimal rewritings: if the TBox is in ELHI, ELH or EL, then the rewriting is a datalog query; if the TBox is in DLlite+, then the rewriting consists of a union of conjunctive queries and a linear datalog program; and if the TBox is in DLliteR, then the rewriting is a union of conjunctive queries. Existing (deductive) database technology can be used to evaluate the rewriting over any ABox.

REQUIEM can be run in three modalities: the Naive modality, which corresponds to the original algorithm; the Full modality, which applies forward/query subsumption and dependency graph pruning, and the Greedy modality, which applies the optimizations of the Full modality plus greedy unfolding (only useful for EL ontologies) (see Using REQUIEM).

REQUIEM is open-source and released under LGPL. All components and source code is included in the downloaded zip file.



Using Eclipse

Currently, REQUIEM can only be used with Eclipse: see Using REQUIEM for more details.


Ian Horrocks, Boris Motik and Héctor Pérez-Urbina


UK Engineering and Physical Sciences Research Council

Key Publications

Héctor Pérez-Urbina, Boris Motik, and Ian Horrocks. Tractable Query Answering and Rewriting under Description Logic Constraints. J. of Applied Logic, 8(2):186-209, 2010. DOI 10.1016/j.jal.2009.09.004
[ bib | .pdf ]