Advances in Bioinformatics / 2012 / Article / Tab 4 / Review Article
Applications of Natural Language Processing in Biodiversity Science Table 4 Performance metrics for the names recognition and morphological character extraction algorithms reviewed. Recall and precision values may not be directly comparable between the different algorithms. NA: not available [
30 ].
Tool Recall Precision Test Corpora Reference TaxonGrab >94% >96% Vol. 1 Birds of the Belgian Congo by Chapin [31 ] FAT 40.2% 84.0% American Seashells by Abbott [32 ] Taxon Finder 54.3% 97.5% American Seashells by Abbott [32 ] Neti Neti 70.5% 98.9% American Seashells by Abbott [32 ] LINNAEUS 94.3% 97.1% LINNAEUS gold standard data set [33 ] Organism Tagger 94.0% 95.0% LINNAEUS gold standard data set [34 ] X-tract NA NA Flora of North America [35 ] Worldwide Botanical Knowledge Base NA NA Flora of China http://wwbota.free.fr/ Terminator NA NA 16 nematode descriptions http://www.math.ucdavis.edu/~milton/genisys/terminator.html MultiFlora mid 60% mid 70% Descriptions of Ranunculus spp. from six Floras http://intranet.cs.man.ac.uk/ai/public/MultiFlora/MF1.html MARTT 98.0% 58.0% Flora of North America and Flora of China [30 ] WHISK 33.33% to 79.65% 72.52% to 100% Flora of North America [13 ] CharaParser 90.0% 91.0% Flora of North America [36 ]
