Ontology Construction – An Iterative and Dynamic Task

To date the World Wide Web consists mainly of loosely structured data. In order to change this situation and make the Semantic Web successful this vast amount of data needs to be given structure and more importantly, meaning. Ontologies are considered a support in this process as they help us explicitly define concepts. A drawback when using ontologies is that they require a significant modelling effort. On the other hand, semantic inter-correspondences as representation of the semantic relationships between two concepts of different information sources can be additional used as a description. However, they lack formal semantics. In this paper, we discuss a new approach in which we combine these two methods. We will discuss our approach with a look at a minimal ontology, generate hypotheses, and verify these with the semantic inter-correspondences. As a side effect we will use the latter to generate an integration formalism, e.g. in form of rules, which we can use for data integration.