Web ontology representation and reasoning via fragments of set theory

TL;DR

This paper introduces a set-theoretic description logic that is highly expressive and decidable, enabling reasoning about semantic web ontologies and rules with NP-complete complexity.

Contribution

It presents a novel set-theoretic description logic, proves its decidability and NP-completeness, and translates SWRL rules into this framework for enhanced reasoning capabilities.

Findings

The logic is as expressive as OWL 2 DL.

Consistency checking is decidable and NP-complete.

SWRL rules can be translated into the set-theoretic framework.

Abstract

In this paper we use results from Computable Set Theory as a means to represent and reason about description logics and rule languages for the semantic web. Specifically, we introduce the description logic $\mathcal{DL}\langle 4LQS^R\rangle(\D)$--admitting features such as min/max cardinality constructs on the left-hand/right-hand side of inclusion axioms, role chain axioms, and datatypes--which turns out to be quite expressive if compared with $\mathcal{SROIQ}(\D)$, the description logic underpinning the Web Ontology Language OWL. Then we show that the consistency problem for $\mathcal{DL}\langle 4LQS^R\rangle(\D)$-knowledge bases is decidable by reducing it, through a suitable translation process, to the satisfiability problem of the stratified fragment $4LQS^R$ of set theory, involving variables of four sorts and a restricted form of quantification. We prove also that, under suitable not very restrictive constraints, the consistency problem for $\mathcal{DL}\langle 4LQS^R\rangle(\D)$-knowledge bases is \textbf{NP}-complete. Finally, we provide a $4LQS^R$-translation of rules belonging to the Semantic Web Rule Language (SWRL).