M5.3 Semantic artefact assessment methodology

Semantic artefacts (i.e., ontologies, vocabularies and SKOS taxonomies, among others) define the structure, guide the construction of, and help validate many existing Knowledge Graphs. In the last years, a number of guidelines have been proposed (Poveda-Villalón et al. 2020; Garijo and Poveda-Villalón 2020; Hugo et al. 2020; Le Franc et al., 2022; Xu et al. 2023) to align semantic artefact best practices against the Findable, Accessible, Interoperable and Reusable principles (FAIR principles) (Wilkinson et al. 2016). Based on these guidelines, new validators and assistants have been developed (Garijo et al. 2021; Amdouni et al. 2022a; 2022b) in order to guide users assessing their own semantic artefacts against the FAIR principles. However, different tests are based on different interpretations of the FAIR principles, resulting in different scores and checks for semantic artefacts. To the best of our knowledge, there is no generic methodology grouping the types of tests to perform in semantic artefacts, in order to map existing assessment efforts in a consistent manner.

In this document, we propose such a methodology. We do so by taking an ontology development perspective, dividing semantic artefacts into smaller parts (their code, content, ontology metadata, etc.) that can be individually assessed at different stages of their development process. We build on the Linked Open Terms (LOT) methodology (Poveda-Villalón et al. 2022), adding a “FAIR assessment” module, and, for each activity, we validate our approach by mapping to two existing semantic artefact FAIR assessment validators: FOOPS! (Garijo et al. 2021) and O’FAIRe (Amdouni et al. 2022a; 2022b). 

The rest of the document outlines our methodology, describes each step in detail, and maps it to existing FAIR principles and guidelines.