Ontology and Instance Matching for the Linked Open Data Cloud
thesisposted on 13.12.2012 by Federico Caimi
In order to distinguish essays and pre-prints from academic theses, we have a separate category. These are often much longer text based documents than a paper.
The linked data paradigm has become a reality as more and more people and organizations publish their data following its principles. It envisions a web made by interlinked datasets that are easy to retrieve, query, and integrate. The main peculiarity of this technology is the presence of links between the data sources as well as the machine-processability of the data, achieved with the use of Semantic Web standards. However, since generating links between those datasets is costly and time-consuming, the need for automatic methods keeps increasing. For this reason ontology matching and instance matching, the fields studying how to automatically match semantic data sources, are being heavily investigated. In this work we present an extension of AgreementMaker, a successful state-of-the-art ontology matching system, to effectively align ontologies and datasets available in the Linked Open Data cloud both at the schema and instance level. To achieve both of the goals, two research directions have been followed: the former is how to improve a general ontology matching system when matching LOD ontologies, while the latter is how to extend it to match instances maximizing the reuse of the components developed for ontology matching.