Ontologies are structural components of modern information systems. The taxonomy, the core of an ontology, is a delicate balance between adequacy considerations, minimal commitments and implementation concerns. However, ontological taxonomies can be quite restrictive and entities that are commonly used in production and services might not find room in a official or de facto standard or ontological system. This mismatch between the company's view and the ontological constraints can limit or even jeoparize the adoption of modern formal ontologies in industry. We study the roots of this problem and individuate a general set of principles to relate the ontology and those non-ontological entities that are yet important for the core business of the company. We then introduce a theoretically sound and formally robust approach to expand a given ontology with new dependency relations, which make available information regarding the non-ontological entities without affecting the consistency of the overall information system. (C) 2014 Published by Elsevier B.V.
An ontological approach for reliable data integration in the industrial domain
Elsevier, Amsterdam , Paesi Bassi
Computers in industry 65 (2014): 1242–1252. doi:10.1016/j.compind.2013.12.010
info:cnr-pdr/source/autori:Borgo, Stefano/titolo:An ontological approach for reliable data integration in the industrial domain/doi:10.1016/j.compind.2013.12.010/rivista:Computers in industry/anno:2014/pagina_da:1242/pagina_a:1252/intervallo_pagine:1242–1