Skip to main content

Table 8 Evaluation results from our system using IG weighting on the BMC dataset

From: Recognizing speculative language in biomedical research articles: a linguistically motivated perspective

Threshold Precision Recall Accuracy F1 score
1 0.48 0.82 0.81 0.60
1.5 0.66 0.74 0.89 0.70
1.75 0.75 0.67 0.90 0.71
2 0.75 0.65 0.90 0.70
2.5 0.88 0.54 0.91 0.67
5 0.97 0.35 0.89 0.52