Presenter
Jacquline Wormstädt (@JacqulineWormstaedt)
Abstract
High-quality, domain-specific metadata is essential for FAIR research data and for ensuring the comprehensibility and reproducibility of research. Although many terminologies and ontologies are available, collecting metadata for a specific research project remains a challenge. Within the AIMS (Applying Interoperable Metadata Standards) project, we have developed an approach for creating domain-specific, RDF-based metadata profiles in the form of SHACL shapes. We use SHACL because it provides an RDF-native, standardized way to express constraints, enabling us to use the same profiles for metadata validation and as a basis for generating metadata forms and subject-specific search interfaces for knowledge graphs.
To make profile modeling accessible to users with limited ontology or SHACL knowledge, we have developed the Metadata Profile Service, a web-based, open-source editor with a graphical user interface for creating metadata profiles. The service supports searching existing terminologies, adding terms to profiles, and defining constraints such as data types, classes, node types, and cardinalities. This allows users without SHACL expertise to create profiles that can be used to create and validate RDF metadata, as well as to display and browse it using the open-source Knowledge Graph Explorer. In addition, the service supports the curation of metadata profiles, enabling communities to agree on de facto standards and recommend them to their members.
In this talk, we will introduce the tools we created around SHACL-based metadata profiles, provide insights into our modeling approach, and discuss the challenges of making RDF- and SHACL-based workflows accessible to users without prior expertise in these technologies.