Skip to main content
. 2019 Dec 3;20:627. doi: 10.1186/s12859-019-3217-3

Table 3.

The effectiveness of the application of full-sentence and SDP features according to the mean F1 scores of 30 different random seeds

Model F1 score
Fulls SDPs Mean SD Min Max
- CNN 43.79 3.39 37.02 51.82
BLSTM - 41.22 14.49 12.82 49.93
BLSTM CNN 45.96 2.87 42.09 52.19

All of the highest scores are highlighted in bold except for the SD