Skip to main content

Table 9 Precision/Recall/F1-score results for gene mention over CRAFT development set: LingPipe with distributed model trained on Genia using indicated evaluation mapping strategies

From: A corpus of full-text journal articles is a robust evaluation tool for revealing differences in performance of biomedical natural language processing tools

 

LingPipe Genia

proteinstar-STAR

LingPipe Genia

proteinstar-GENESTAR

LingPipe Genia

protein-STAR

 

Prec

Recall

F1

Prec

Recall

F1

Prec

Recall

F1

strict

0.29

0.38

0.33

0.10

0.25

0.14

0.30

0.37

0.33

shared

0.35

0.47

0.40

0.14

0.34

0.20

0.36

0.45

0.40

subspan

0.36

0.48

0.41

0.14

0.36

0.20

0.37

0.47

0.41

overlap

0.36

0.48

0.41

0.14

0.36

0.20

0.37

0.47

0.41