Figure 1.

Percentage of articles annotated with different numbers of species in various data sets. Training (32) refers to the human annotations on the 32 articles in the training set. Test (50) and Test (507) refer to the gold standard and silver standard annotations on the 50 and 507 articles in the test set, respectively.

Lu et al. BMC Bioinformatics 2011 12(Suppl 8):S2   doi:10.1186/1471-2105-12-S8-S2