Interoperability of Semantic Annotations

Abstract

Abstract The Semantic Web is the new generation World Wide Web. It extends the Web by giving information a well defined meaning, allowing it to be pro- cessed by machines. This vision is going to become reality thanks to a set of technologies which have been specified and maintained by the World Wide Web Consortium (W3C), and more and more research efforts from the industry and the academia. Therefore, the basis for the Semantic Web are computer- understandable descriptions of resources. We can create such descriptions by annotating resources with metadata, resulting in "annotations" about that re- source. Semantic annotation is the creation of metadata and relations between them with the task of defining new methods of access to information and en- riching the potentialities of the ones already existent. The main goal is to have information on the Web, defined in such a way that its meaning could be explicitly interpreted also by automatic systems, not just by humans. There is huge amount of interesting and important information represented through semantic annotations, but there are still a lot of different formalisms showing a lack of standardization and a consequent need of interoperability. This growing need of interoperability in this field convinces us to extend our first proposal, strictly related to database models, in order to address also semantic annotations. Our proposal, mainly based on Model Management techniques, focuses on the problem of translating schemas and data between Semantic Web data models and the integration of those models with databases models that are a more rigid and well-defined structure. In this work we underline the main concepts of our approach discussing a proposal for the implementation of the model management operator Model- Gen, which translates schemas from one model to another focusing on semantic annotation context. The approach expresses the translation as Datalog rules and exposes the source and target of the translation in a generic relational dic- tionary. This makes the translation transparent, easy to customize and model- independent. The proposal includes automatic generation of translations as composition of basic steps.