Semantic interface Requests to a complex industrial software

Unlike web applications designed to be accessed by standardized interfaces, in many cases industrial software solutions have proprietary interfaces with limited functionality and flexibility. We consider how to obtain data from industrial software solutions via sematic interfaces. While most other papers examine descriptive query languages at the database level, here the access to an application with a complex data structure is only possible at a higher level of abstraction using a procedural programming interface. Materialized and virtual integration architectures for Semantic Web were analyzed in the context of accessibility limitations through a programming interface. It turns out that the materialized integration is preferable. In addition, the added value of an integration often increases because other software applications can be semantically linked and combined queries are possible. Finally it also became clear that the source data structure must meet certain requirements so that it is possible to automatically generate an ontology for inferring additional knowledge.