Following the Guidelines on Open Access to Scientific Publications and Research Data in Horizon 2020, the European Commission defines the research data as factual or numerical information, collected to be examined and considered as the basis of reasoning, discussion or calculation.
The data can be statistics, results of experiments, measurements, observations resulting from the field study, surveys or interviews and images.
Research data FAIR is a set of principles that make data findable, accessible, interoperable, and reusable.
How to make the data FAIR?
- (F1) Metadata must have a unique global and persistent identifier (handle, DOI, ...)
- (F2) The data is described with rich metadata
- (F3) Metadata is recorded or indexed in a search engine
- (F4) There is specific metadata for the permanent identifier
- (A1) Metadata can be retrieved using standardized, open, free, and universally implementable communication protocols. When necessary for reasons of privacy, security or commercial interests, the protocol should allow for a system of authentication and authorization of data.
- (A2) Metadata must be accessible even when data is no longer available
- (I1) Metadata uses a formal, accessible, shared, and widely applicable language for the representation of knowledge.
- (I2) Metadata uses a FAIR vocabulary (controlled vocabularies)
- (I3) Metadata includes qualified references to other metadata (publications and related materials)
- ( Metadata is described with a plurality of precise and relevant attributes, including its provenance and a clear and accessible license to use.