Skip to main content

Table 10 Performance of various semi-supervised classifiers in identifying argument roles.

From: Semi-supervised learning of causal relations in biomedical scientific discourse

Classifier

P

R

F1

Mihăilă et al. [38]

85.25%

83.55%

84.35%

Naïve Bayes

70.45%

80.05%

74.94%

SVM

82.50%

80.05%

81.25%

JRip

84.65%

80.90%

82.73%

J48

83.10%

79.20%

81.10%

RandFor

79.85%

74.20%

76.92%

Vote

84.55%

83.05%

83.79%