Streaming semantic translations

As the Linked Data paradigm is embraced by a growing number of data providers (and consumers), the count of publicly available, semantically-annotated resources is systematically increasing. Here, elasticity and extensibility of semantics, enmeshed in Linked Data principles, and technologies supporting them, starts to attract application developers. In this context, multitude of developed ontologies motivates (and, in some cases, forces) work devoted to ontology “reconciliation”, including alignment and merging, as well as supports pursuing rules of ontology engineering, such as modularity and reuse. At the same time, as uptake of semantic technologies grows, semantically-annotated resources are used not only in static data storage systems (where they appear “in bulk” and are accessed through some query mechanism), but also as separate, dynamically created, transmitted and processed entities. In the field of the Internet of Things, this is exemplified in content of messages exchanged, either between internal components of IoT platforms (within the middleware), or between applications built on top of each other. Platforms such as Open-IoT, UniversAAL, or VITAL, use serializations of RDF to transmit messages, while other platforms, such as FIWARE and OneM2M support ontologies, even if they don't use RDF within exchanged messages. However, one of important unsolved issues remains - “how to facilitate understanding of messages exchanged between artifacts founded on different ontologies” (i.e. establish semantic interoperability). In this paper, we focus on semantic translation applied in the context of “continuous” communication between IoT artifacts. Our approach is rooted in semantic translation, based on ontology alignments, and utilizes scalable streaming infrastructure that supports diverse communication topologies and scenarios. It also forms foundation for the design and implementation of the Inter Platform Semantic Mediator (IPSM), which is one of key components to establish interoperability in the INTER-IoT project.