Search

The “FAIR” principles define a set of characteristics data and their metadata should respect in order to be Findable, Accessible, Interoperable, and Reusable. Per principle I2, an ontology, and more generally a semantic resource, should also be FAIR. Ontology-specific criteria begin to emerge, but there is still no mechanism to assess the degree of implementation of these principles. We propose an automatic evaluation method of the level of "FAIRness" of an ontology which is based on a rich and structured representation of the metadata. We rely on the MOD metadata model and have developed a prototype for AgroPortal, a repository of semantic resources and ontologies in agronomy., Les principes "FAIR" définissent un ensemble de caractéristiques que les données et leurs métadonnées devraient présenter pour être Faciles à trouver, Accessibles, Interopérables et Réutilisables. Également, suivant le principe I2, une ontologie, et plus généralement une ressource sémantique, devrait aussi être "FAIR". Des critères spécifiques aux ontologies commencent à apparaître, mais il n'existe toujours pas de mécanismes pour évaluer le degré de mise en oeuvre de ces principes. Nous proposons une approche d'évaluation automatique du niveau de "FAIRness" d'une ontologie qui se base sur une représentation riche et structurée des métadonnées. Nous nous reposons sur le modèle de métadonnées MOD et avons développé un prototype pour AgroPortal, un portail de ressources sémantiques et d'ontologies en agronomie.

International audience; As any resources, ontologies, thesaurus, vocabularies and terminologies need to be described with relevant metadata to facilitate their identification, selection and reuse. For ontologies to be FAIR, there is a need for metadata authoring guidelines and for harmonization of existing metadata vocabularies –taken independently none of them can completely describe an ontology. Ontology libraries and repositories also have to play an important role. Indeed, some metadata properties are intrinsic to the ontology (name, license, description); other information, such as community feedbacks, or relations to other ontologies are typically information that an ontology library shall capture, populate and consolidate to facilitate the processes of identifying and selecting the right ontology(ies) to use. We have studied ontology metadata practices by: (i) analyzing metadata annotations of 805 ontologies; (ii) reviewing the most standard and relevant vocabularies (23 totals) currently available to describe metadata for ontologies (such as Dublin Core, Ontology Metadata Vocabulary, VoID, etc.); (iii) comparing different metadata implementation in multiple ontology libraries or repositories. We have then built a new metadata model for our AgroPortal vocabulary and ontology repository, a platform dedicated to agronomy based on the NCBO BioPortal technology. AgroPortal now recognizes 346 properties from existing metadata vocabularies that could be used to describe different aspects of ontologies: intrinsic descriptions, people, date, relations, content, metrics, community, administration, and access. We use them to populate an internal model of 127 properties implemented in the portal and harmonized for all the ontologies. We –and AgroPortal's users– have spent a significant amount of time to edit and curate the metadata of the ontologies to offer a better synthetized and harmonized information and enable new ontology identification features. Our goal was also to facilitate the comprehension of the agronomical ontology landscape by displaying diagrams and charts about all the ontologies on the portal. We have evaluated our work with a user appreciation survey which confirms the new features are indeed relevant and helpful to ease the processes of identification and selection of ontologies. This paper presents how to harness the potential of a complete and unified metadata model with dedicated features in an ontology repository, however the new AgroPortal's model is not a new vocabulary as it relies on pre-existing ones. A generalization of this work is studied in a community driven standardization effort in the context of the RDA Vocabulary and Semantic Services Interest Group.

National audience; Scientific communities are using an increasing number of ontologies. Repositories make them available, like the NCBO BioPortal which currently hosts more than 500 biomedical ontologies. Now the question is how to find the ontology we need? One solution is to describe each ontology with appropriate metadata. However, none of the existing metadata vocabularies can completely meet this need if taken independently. We have reviewed a large number of vocabularies, such as Dublin Core, OMV, DCAT, or VOID, as well as the properties implemented by common ontology repositories. We then listed those properties into a simplified model of 124 properties. We present a few examples of use of these properties within the AgroPortal, an ontology repository for agronomy, and explain how the portal handles these properties to facilitate ontology description and selection.; Les communautés scientifiques utilisent un nombre croissant d'ontologies. Pour les mettre à disposition, il existe des portails d'ontologies, à l'exemple du NCBO BioPortal qui regroupe actuellement plus de 500 ontologies biomédicales. Mais face à cette avalanche de ressources, comment trouver l'ontologie qui répondra à nos besoins ? Une solution consiste à décrire chaque ontologie avec des métadonnées appropriées. Or, il n'existe pas à ce jour de vocabulaire de métadonnées suffisamment exhaustif pour répondre à ce besoin. Nous avons passé en revue un grand nombre de vocabulaires, tels que Dublin Core, OMV, DCAT ou VOID ainsi que les propriétés implémentées par les portails d'ontologies les plus courant. Nous en avons produit un modèle simplifié composé de 124 propriétés. Nous présentons ici quelques exemples d'utilisation de ces propriétés dans AgroPortal, un portail d'ontologies dédié à l'agronomie, et nous expliquons comment elles sont gérées et utilisées pour la description et l'identification d'ontologies.

International audience; Similarly to what happens in biomedicine, communities engaged in agronomic research need to access specific sets of ontologies for data annotation and integration. For instance, it has been established that the scientific challenges in plant breeding have switched from genetics to phenotyping and that standard traits/phenotypes vocabularies are necessary to facilitate breeder’s data integration and comparison. In parallel of very specific crop dictionaries, important organizations have produced large reference vocabularies such as AGROVOC (Food and Agriculture Organization), NAL Thesaurus (National Agricultural Library) or the CAB Thesaurus (Centre for Agricultural Bioscience International). The more ontologies are being produced in the domain, the more the need to create, store and retrieve alignments between those ontologies become important. In fact, there exists a need of a one-stop-shop for agronomical, environmental and food related ontologies enabling to identify and select an ontology for a specific task as well as offering generic services to exploit them in search, annotation or other scientific data management processes