Biodiversity Knowledge Graphs: Time to move up a gear!

Harnessing worldwide biodiversity data requires integrating myriad pieces of information, often sparse and incomplete, into a global, coherent space. To do so, projects like the Global Biodiversity Information Facility, Catalog Life Encyclopedia have set up platforms that gather, consolidate, centralize billions records from multiple sources. This approach lowers entry barrier for scientists willing to consume aggregated but tends build silos hamper cross-platform interoperability. The Web Data embodies different underpinned by Linked Open (LOD) principles (Heath Bizer 2011). These bring about building large, distributed, cross-domain knowledge graph (KG), wherein description relies on vocabularies with shared, formal, machine-processable semantics. So far however, little been published this way. Early efforts focused primarily taxonomic registers, such as NCBI, VTO AGROVOC. More recent started paving way publication more diverse KGs (Page 2019, Penev et al. Michel 2017). Today, we believe it is time producers join in start publishing connected spanning much broader domains, beyond just registers. In talk, wish present an on-going endeavor line vision. previous work, TAXREF-LD (Michel 2017), LOD representation French register developed maintained National Museum Natural History. We modeled nomenclatural information thesaurus scientific names, ontology classes denoting taxa, additional ranks vernacular names. Recently, extended scope represent interlink various geographic locations, species interactions, development stages, trophic levels, well conservation, biogeographic, legal status (regulations, protections, etc.). put specific effort working out model accurately accounts semantics while respecting engineering practices. For instance, common design shortcoming attach all properties taxon. rightful choice some name or conservation status, actually pertain biological individuals themselves, e.g. habitat level, should better be attched class members. With presentation advance discussion integration scenarios based graphs stakeholders.