Verification of Global and Local as View Data Integration in an Object Data Model

Semantic data integration is a challenging field of research that is required for uniform access, interpretation, and further utilization of heterogeneous data sources with increasing volume, variety, and complexity. Over the last twenty years, many technologies and systems have been developed to support semantic data integration. However, verification of data integration meaning the formal proof that semantics of data and behavior are preserved during data integration is still a challenging problem. This paper concentrates on data integration using views, namely on Global and Local as View (GLAV) data integration in an object data model. An algorithm for the application of a specification language based on first-order logic and set theory and automated provers for data integration verification is proposed. An advantage of object data models is the combining of data structures and methods for their manipulation in classes. This allows us to integrate both structured relational and NoSQL databases as well as behavior-driven data sources like web services. The algorithm is illustrated via a use case in the land use management domain.