Automating the transformation of DMN decision tables into the SPARQL CONSTRUCT query form

Abstract

This master thesis tries to define a bridge which links decisions expressed in DMN 1.1 and rules written in SPARQL queries. Indeed, DMN 1.1 is a human-understandable standard, specified by the Object Management Group, which allows users to specify decisions. Essential for DMN are decision tables. Decision tables are easy to fill in and understand; indeed, elements of decision tables are expressed in a very simple way. On the other hand, ontologies exist, which hold all the data and concepts. SPARQL queries are machine-understandable and are defined by a strict syntax, which is in contrast with the decision tables representation. The execution of SPARQL queries can create an inference, which allows a reasoning able to bring new insights starting from a pre-existent information base. This study implements a solution for the transformation of decision tables in SPARQL queries, through an in-depth analysis of the semantics of both sides. The development of an artefact is implied, which automates the transformation from decision tables to SPARQL queries, ready to be executed over ontologies.