Skip to main content
. 2023 May 3;13:7173. doi: 10.1038/s41598-023-34294-6

Table 10.

All experiments have been evaluated on the test set.

Model type F1-score (%)
microsoft (med-bert) 91
biobert (https://github.com/dmis-lab/biobert/)+characterBert 92.7
bertConfig+CharacterBert 93.66
bertConfig+CharacterBert+focalLS 93.45
microsoft+focalLs 91.05
microsoft+characterBert 91.28

Significant values are in [bold].