Skip to main content

Table 7 Evaluation of the system on the two annotators’ test sets. We reproduce IAA from Table 3 for comparison

From: Text mining brain imaging reports

 

Precision

Recall

F1

IAA F1

Entities

    

Annotator 1 test set

94.63

96.37

95.49

96.96

Annotator 2 test set

97.21

96.50

96.86

 

Negation

    

Annotator 1 test set

93.54

95.30

94.41

96.46

Annotator 2 test set

96.35

95.66

96.01

 

Relations

    

Annotator 1 test set

97.32

99.24

98.27

95.84

Annotator 2 test set

95.47

97.61

96.53

 

Labels

    

Annotator 1 test set

94.94

97.88

96.39

94.02

Annotator 2 test set

92.70

92.52

92.61

Â