Skip to main content
. 2018 Jun 13;19(Suppl 8):209. doi: 10.1186/s12859-018-2195-1

Table 5.

Results obtained for attentive pooling CNN on the test dataset with negative filtering

Classes P R F1
Advise 78.74% 61.99% 69.37%
Effect 58.29% 57.14% 57.71%
Int 79.07% 35.42% 48.92%
Mechanism 60.75% 54.03% 57.19%
Overall 64.42% 55.14% 59.42%