Skip to main content
Fig. 3 | BMC Bioinformatics

Fig. 3

From: Bio-SimVerb and Bio-SimLex: wide-coverage evaluation sets of word similarity in biomedicine

Fig. 3

Subset-based evaluation for Bio-SimVerb (y axis unit: ρ), where subsets are created based on the word’s number of unique Broad Subject Terms. A word can have multiple Broad Subject terms when it appears in journals of different areas in biomedicine. To be included in each group, it is required that both words in a pair are contained in the same Subject Term interval (x axis)

Back to article page