2012 | OriginalPaper | Buchkapitel
OtO Matching System: A Multi-strategy Approach to Instance Matching
verfasst von : Evangelia Daskalaki, Dimitris Plexousakis
Erschienen in: Advanced Information Systems Engineering
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
In this paper we describe an ontology to ontology (OtO) matching system which implements a novel instance matching algorithm. The proposed multi-strategy matching system is domain independent and fully customizable at any level. It optimizes the instance matching process by leveraging (a) the rich semantic knowledge from the schema matching results, (b) the implicit knowledge of the domain expert by capturing the identification power of the properties and (c) the probability estimation of the result’s validity, in order to accurately detect the ontology instances that represent the same real-world entity. Furthermore we evaluate the system with the ISLab Instance Matching Benchmark in the Ontology Alignment Evaluation Initiative 2009 campaign and report the results.