Challenges of General Adaptation Framework for Industrial Semantic Web

Integration of heterogeneous applications and data sources into an interoperable system is one of the most relevant challenges for many knowledge-based corporations nowadays. Development of a global environment, which would support knowledge transfer from human experts to automated Web-services which are able to learn, is a very profit-promising and challenging task. The domain of industrial maintenance is not an exception. This paper outlines in details an approach for adaptation of heterogeneous Web resources into a unified environment as a first step towards interoperability of smart industrial resources, where distributed human experts and learning Web-services are utilized by various devices for self-monitoring and self-diagnostics. The proposed General Adaptation Framework utilizes a potential of the Semantic Web technology and primarily focuses on the aspect of a semantic adaptation (or mediation) of existing widely used models of data representation to RDF-based semantically rich format. To perform the semantic adaptation of industrial resources the approach of two-stage transformation (syntactical and semantic) is elaborated and implemented for monitoring of a concrete industrial device with underlying XML-based data representation model as a use case.