In the first pilot sprint, the Connectome project partners focused on defining the needs of scientific communities in terms of how to increase the accessibility and reuse of open linked data for research.
For the prototype of the Research Data Connectome, the project partners developed an ontology as the basis for the Connectome knowledge graph. This ontology was defined by transforming user problems, identified through 21 interviews with researchers, into so-called Ontology Competency Questions. These Competency Questions defined the capabilities for the Connectome Knowledge Graph – or in other words: what user’s can do with it through services (e.g. retrieving information of research peers, recommendations on relevant publications and datasets).
The elaborated RESCS ontology for the Connectome Knowledge Graph is derived from, and extends upon, best practices such as schema.org. This was already anticipated during the design phase for the Connectome architecture.
In addition to the above work, we compared our ontology with Best Practices from the research domain (e.g. OpenAIRE Research Graph) to select those features that covered the needs of the researchers most.
The Connectome aims to contribute to increasing interdisciplinarity in research by providing a knowledge graph which links the metadata of many different relevant data providers.
We used the RESCS Ontology for two purposes within the Connectome Pilot. First, to prototype the Connectome Linked Data Pipeline that harvests, extracts and harmonises metadata from various data providers (such as DaSCH / University of Basle, FORS, or OpenAIRE) into the metadata-structure of the RESCS ontology using the Blue Brain Nexus Forge features. Second, we deposited a machine-readable version of the RESCS Ontology in Blue Brain Nexus platform running on SWITCHengines, in order to allow import and validation of the metadata coming from the Connectome Linked Data Pipeline. Moreover, we documented our RESCS ontology using a web-based-interface on the domain rescs.org.
The RESCS ontology shows already in its prototype version how the knowledge graph can create added values for different stakeholders. Our aim for 2021 is, to create an open community of data- and service providers, as well as scientific users, to govern and extend the RESCS ontology (e.g. to increase data linkages or discipline specificities). Our focus is the enablement of Service Providers to add value to researchers by re-using the metadata linked through the Connectome.
A knowledge graph can be seen as a structured system of interlinked entities (e.g. data, publications, persons, etc.) on the basis of a specific ontology. A knowledge graph enables both the discovery of regularities and the appreciation of relationships that would otherwise remain opaque. By putting data in a knowledge graph, a unifying horizon of meaning for the interlinked entities is realised, and new discovery and research paths, as well as new kind of interdisciplinary questions, become possible.