Matching material citations to occurrences: extending the biodiversity knowledge graph

Presenter: Donat Agosti

Scholarly publications are the channel to communicate research findings based on the analysis of specimens and their traits. In an open science world, the link to the specimens together with the description of the methods involved should enable us to reproduce the results. Increasingly, taxonomic publications provide a section with structured or even semantically enhanced tables, including persistent digital identifiers of the specimens. These publications can be processed, reformatted as datasets of material citations that can be reused, for example by the Global Biodiversity Information Facility (GBIF). In many cases these are the only representations of species in GBIF, but in other cases they complement specimens that have been submitted as preserved specimens. In this case, the presence of a material citation provides in return access to the knowledge about the specimen, or in other words, extends the biodiversity knowledge graph from a specimen to what is stated in the respective hosting taxonomic treatment, often far beyond by all the links that are embedded in it. Matching can be done automatically but in many cases depends on human curation. For this case a matching service developed with the Swiss eBioDiv and the European funded BiCIKL projects will be explained.