Using the ontology paradigm to integrate information systems

Abstract

Ontology based websites are one possible implementation of the Semantic Web. There are several languages for ontology specification: RDF, OWL, Topic Maps. Topic Maps follow a structure formally specified what makes them a good choice for semantic website specification. The process of ontology development based in topic maps is complex, time consuming, and it requires a lot of human and financial resources, because they can have a lot of topics and associations, and the number of information resources can be very large. To overcome this problem a new environment is proposed, Oveia. Oveia is composed by four components which have relevant contributions to the Semantic Web area. This paper describes these components in detail. Two components representing a metadata extractor: heterogeneous data integration (through XSDS specifications) and an homogeneous intermediate data representation for the extracted metadata (datasets). The Ontology builder who builds an ontology from metadata stored in a set of datasets (construction rules are specified in a new domain specific language: XS4TM). The Ontology builder stores the result in XTM files or in relational databases according to the Topic Map structure. Finally, Ulisses, the navigational component, generates web interfaces through which is possible to move inside the topic map and among information resources.