Incorporating rich background knowledge for gene named entity classification and recognition

BMC Bioinformatics

Table 1 Statistical information and NER performance of dictionaries

Dictionary	#of entries	Coverage	Precision	Recall	F-score
BioThesaurus	4,480,469	36.78%	15.36	77.21	25.62
ABGene lexicon	1,101,716	35.98%	31.54	53.58	39.71
Combined	5,522,822	54.32%	16.20	82.59	27.09
Combined+varients	10,034,696	65.68%	7.32	79.26	13.40

The rightmost three columns show the recognition performances of dictionaries on BioCreative 2 test corpus using maximum match method.

ISSN: 1471-2105