Skip to main content

Table 9 Performance of various semi-supervised classifiers in identifying dependent (DA) and independent (IA) argument spans.

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

Classifier P R F1
Mihăilă et al. [38] 74.18% 88.98% 80.91%
CRF 84.52% 79.58% 81.98%
SVM 75.85% 77.95% 76.89%
Random Forest 76.95% 76.50% 76.72%
Naïve Bayes 63.30% 67.35% 65.26%