Abstract:
OWL RL has been recently proposed by the W3C consortium as a fragment of OWL 2 for which reasoning can be done efficiently. In this context, we have studied how to implement some heuristics within this profile to enhance the response time of the materialization process. Basically, our work goes towards the use of SPARL 1.1 as query and reasoner engines. Several strategies were proposed and compared for this entailment regime using different data stores. The results of this study shows that rules dependency can have a great impact when the set of applied rules of the regime are not strongly dependent. The approach presented in this paper can be easily extended and applied to other OWL profiles or RDF entailment regimes.
