Data Manangement Handbook for S-ENDA partners

Vocabulary

An important part of ensuring the FAIRness of the data is to adopt standardized and controlled vocabularies to describe the data and metadata. Controlled vocabularies are controlled lists of terms used to organize information. They require the use of predefined terms to ensure consistency and accuracy in documenting data for easy dataset discovery. The vocabularies themselves shall also follow the FAIR principles, highlighted by the I2 (Interoperability 2) principle. In practice, this means that vocabularies should ideally be:

• Findable -> registered (indexed, listed) in a vocabulary service
• Accessible -> available on the web, downloadable
• Interoperable -> encoded in a standard representation, such as the Web Ontology Language (OWL) or Simple Knowledge Organization System (SKOS) and domain-specific extensions
• Reusable -> licensed and maintained, ideally with an open license (Source: FAIR vocabularies)

The formalised vocabularies agreed upon during the project aim to facilitate data search in ADC portal across S-ENDA partner data centres. Ideally, each institute provides FAIR compliant machine-to-machine access point to harvest vocabularies, enabling mapping of internal vocabulary to common reference vocabulary (e.g. REST API or SPARQL endpoints for vocabulary servers).

The table below shows the current domain-specific reference vocabularies together with vocabulary servers for each institute.