Ontology Evolution in Data Integration1

Ontology Dynamics 2010.

力学;计算机科学

Haridimos Kondylakis ; Dimitris Plexousakis ; Yannis Tzitzikas

Others  :  http://ceur-ws.org/Vol-651/paper4.pdf
PID  :  42049

来源: CEUR

【 摘 要 】

Ontologies are becoming more and more important in data integration. An important problem when dealing with ontologies is the fact thatthey are living artefacts and subject to change. When ontologies evolve, thechanges should somehow be rendered and used by the pre-existing dataintegration systems. In most of these systems, when ontologies change theirrelations with the data sources i.e. the mappings, are recreated manually, aprocess which is known to be error-prone and time-consuming. In this paper,we provide a solution that allows query answering under evolving ontologieswithout mapping redefinition. In order to achieve that, we present a module thatenables ontology evolution over traditional ontology-based data integrationsystems. This module gets as input the different ontology versions and the userquery, and answers queries over data integration systems that use differentontology versions. We identify the problems in such a setting and we provideefficient, intuitive solutions. We prove that our approach imposes only a smalloverhead over traditional query rewriting algorithms and it is modular andscalable. Finally, we show that it can greatly reduce human effort spent since continuous mapping redefinition on evolving ontologies is no longer necessary.