Affiliations: [a] Department of Mathematics and Computer Science, University of Calabria, Italy
| [b] DLV System Srl, Piazza Vermicelli, Polo Tecnologico UniCal, Rende, Italy
Corresponding author: Francesco Calimeri, Department ofMathematics and Computer Science, University of Calabria, Via Bucci, Cubo 30B, I-87036 Rende, Italy. Tel.: +39 0984 496478; Fax: +39 0984 496410; E-mail: [email protected].
Abstract: DLV is a powerful system for Knowledge Representation and Reasoning which supports Answer Set Programming (ASP) – a logic-based programming paradigm for solving problems in a fully declarative way. DLV is currently widely used in academy, and, importantly, it has been fruitfully employed in many relevant industrial applications. Similarly to the other main-stream ASP systems, while processing an input program, in a first phase of the computation DLV eliminates the variables, thus generating a ground program which is semantically equivalent to the original one, but significantly smaller than the Herbrand Instantiation, in general. This phase, called ‘grounding’, plays a key role for the successful deployment in real-world contexts. In this work we present ℐ-DLV, a brand new version of the intelligent grounder of DLV . While relying on the solid theoretical foundations of its predecessor, it has been completely redesigned and re-engineered, both in algorithms and data structures; it now features full support to ASP-Core-2 standard language, increased flexibility and customizability, significantly improved performance, and an extensible design that eases the incorporation of language updates and optimization techniques. ℐ-DLV results in a stable and efficient ASP instantiator, that turns out to be a full-fledged deductive database system. We describe here the main features of ℐ-DLV and present the results of proper experimental activities for assessing both its applicability and performance.
Keywords: Answer Set Programming, DLV, Artificial Intelligence, Knowledge Representation and Reasoning, Grounding, Instantiation, Deductive Database Systems