Graph-based Semantic Evolution for Context Information Management Platforms

Context Information Management (CIM) platforms have tended to rely on mostly hierarchical information models with weak semantics and predefined APIs, falling short of the requirements for interoperation with other platforms and flexibility of data utilization and federation. In order to improve cross-domain federation capabilities for existing CIM platforms, we propose an evolution framework by combining the “property graph” information model with RDF-based semantic modeling. We describe this evolution for what concerns: context information modeling, context information representation, query patterns and architecture. For each of these aspects, we introduce the problem space, the evolution strategy and provide examples based on existing popular CIM platforms. This evolution framework is designed to meet the requirement of backward compatibility to existing CIM platforms, while bringing graph-based semantic evolution based on the property graph model for CIM interoperability and cross-domain federation. This framework is currently evolved by the ETSI Industry Specification Group on “cross-cutting Context Information Management” (ETSI ISG CIM) for standardization.