issues
search
bst-mug
/
n2c2
Support code for participation at the 2018 n2c2 Shared-Task Track 1
https://n2c2.dbmi.hms.harvard.edu
Apache License 2.0
6
stars
4
forks
source link
[NN] Saliency analysis
#88
Closed
michelole
closed
5 years ago
michelole
commented
5 years ago
Idea to analyze if NN is learning something useful:
train a model M
for each label L do:
for each sentence S in training do:
classify S for L using the model M
store probability P
sort sentences by P
analyze the sentence S with highest P for L (according to M)
Idea to analyze if NN is learning something useful: