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

Table 6.

Results obtained for the combination of max-pooling and attentive pooling CNN on the test dataset corpus with negative filtering

Classes P R F1
Advise 79.23% 65.61% 71.78%
Effect 65.28% 61.62% 63.40%
Int 80.49% 34.38% 48.18%
Mechanism 69.23% 60.40% 64.52%
Overall 70.40% 59.47% 64.47%