Skip to main content
. 2019 Jan 16;42(1):147–156. doi: 10.1007/s40264-018-0763-y

Table 5.

Performance metrics of the relation extraction model on the final 213 evaluation documents

Relation category Precision Recall F1
Dosage 95.7 96.2 96.0
Frequency 97.1 92.3 94.7
Route 96.1 92.1 94.1
Severity 91.1 96.2 93.6
Duration 93.7 91.2 92.4
Reason 78.0 73.9 75.8
Adverse 78.7 68.3 73.1
Overall micro 90.3 85.9 88.1