INSIDE: Semantic Interoperability in Engineering Data Integration
INSIDE is presented, a system that enables Semantic Interoperability for Engineering Data Integration and represents queries to one or multiple databases through the concept of data services, where each service is defined using an ontology.
Abstract
: One of the critical problems in the industry is data integration. Data is generated continuously and on a large scale and is persisted using different formats, software, and terminologies. Integrating multiple databases belonging to different information systems can provide a unified data view and a better understanding of the data. With that in mind, this paper present INSIDE, a system that enables Semantic Interoperability for Engineering Data Integration. INSIDE represents queries to one or multiple databases through the concept of data services, where each service is defined using an ontology. Data services can conceptually represent the commitments and claims between service providers (databases) and service customers (users) along with the service lifecycle (the process of querying, integrating, and delivering data). The contributions of this paper are the following: (1) Use of formal mechanisms for semantic data representation with the connection with the international community; (2) A conceptual model for a distributed system based on ontologies for querying and manipulating data from multiple data sources; (3) An implementation of this model, called INSIDE, developed on top of Apache Spark; and (4) An experimental evaluation of the service composition strategy of INSIDE for data integration of multiple data sources using a real-world scenario.