The Australian Taxonomy

Community Directory

Bloodhound is an application developed and maintained by David P. Shorthouse, that links contributors to taxonomy (collectors and determiners of specimens) to their contribution (specimen records and publications that use those records). It was launched in August 2018 as a submission to the annual Ebbe Nielsen Challenge.

Bloodhound uses specimen data downloaded every two weeks (approximately 204M specimen records per download) from the Global Biodiversity Information Facility (GBIF). In addition, GBIF cited data download packages are downloaded every two weeks, extracted and linked to attributed specimens records.

Names of collectors and determiners are parsed from the specimen records and cleaned using the test-driven dwc_agent ruby gem available for free integration in other projects. Names are analysed using a graph theory method outlined by R.D.M. Page to find name variants that are likely to represent the same person; these are then used to help expand the search for candidate specimens, presented in order of greatest to least probable.

Authentication is provided by ORCID. Anyone who has an ORCID profile can use Bloodhound to search for their specimens and determinations and to view a summary of their contributions to global taxonomy. Biographical details for collectors and determiners are sourced from ORCID and wikidata

Bloodhound also connect with Zenodo using ORCID credentials. Once a set-it-and-forget-it connection is made, Bloodhound pushes specimen data into this industry-recognized, stable, long-term archive and mints a new DataCite DOI. The Zenodo token is cached in Bloodhound and a new version of the specimen data is pushed to the archive A DataCite DOI badge is set on the Bloodhound profile page along with a citation for professional resumes. 

You can read more detailed notes on how Bloodhound works on the How it works page

bloodhound.png
TaxonomyAustralia_logo_inverted.png