Create Jupyter Notebook to compare NER/WoRMS with GBIF

[amandawhitmire]

Our current pipeline for identifying species occurrences is to use the WoRMS taxonomy as an entity set for NER analysis of the student papers. GBIF offers a tool to check species names (https://www.gbif.org/tools/species-lookup). Could we use something like this to improve our accuracy in identifying species in our text?

A first step might be to compare the initial output of our NER process, a list of genus and species names, with the species that exist in GBIF. Is everything we found via WoRMS in GBIF? If so, is there a different pipeline for pulling species names from papers via GBIF that would be more accurate? Consider scenarios when the name is misspelled or the OCR botched a few letters. GBIF offers from fuzzy logic in string matching.