target word 0.002617563
current word 0.002551397
next word 0.002483184
oov word 0.0024239229999999997
single word 0.002420348
individual word 0.002394534
word sense 0.002386765
output features 0.002121734
training data 0.0020978200000000002
input features 0.0019474210000000002
learning data 0.001864791
such words 0.001848564
training time 0.001767347
certain features 0.001764874
recoverable features 0.0016918050000000002
unnecessary features 0.001678402
definiteness features 0.001678402
layer network 0.00166525
test data 0.001644589
input data 0.001556691
output layer 0.001553378
hidden layer 0.0015488189999999999
training corpus 0.0015449840000000001
pos tagger 0.0015431709999999999
unknown words 0.001516376
features 0.00146176
learning algorithm 0.0014431180000000002
training sentences 0.001434361
training speed 0.0014100330000000002
input layer 0.001379065
neural network 0.001376139
several network 0.0013708729999999999
training procedure 0.0013705520000000001
high training 0.001367022
training example 0.0013532140000000002
network test 0.001345405
training process 0.0013386820000000001
test accuracy 0.001337214
previous layer 0.001321262
pos tags 0.001314184
corresponding pos 0.0013091819999999999
training iteration 0.001303453
optimal training 0.0013029920000000002
serious data 0.001302979
data sparseness 0.001300789
layer size 0.001296755
unseen data 0.0012926699999999999
training iterations 0.001276142
other network 0.0012737769999999998
set accuracy 0.001273409
network approach 0.001258684
output vector 0.001240379
msd tagging 0.00121626
words 0.00121619
first network 0.001192465
accuracy figure 0.001185993
layer configuration 0.001180357
output msd 0.001168762
network architecture 0.001164427
size time 0.001143908
layer train 0.00113919
baseline accuracy 0.001139129
msd tagger 0.001126279
layer configurations 0.001114872
accuracy cross 0.001110301
different approach 0.0011077770000000001
test corpus 0.001091753
tagging speed 0.001090715
test set 0.0010833129999999998
system language 0.001076007
language models 0.001068158
network configuration 0.001058799
hidden size 0.001058766
custom network 0.001051824
tagging procedure 0.001051234
accuracy drops 0.0010506320000000001
information extraction 0.001049146
maximum accuracy 0.001048425
msd language 0.001048179
single network 0.0010445139999999999
accuracy rate 0.001037174
unit network 0.001030238
different methods 0.001029373
average accuracy 0.001027704
training 0.00102679
validation accuracy 0.001025564
activation function 0.001024675
individual network 0.0010187
other msd 0.001010719
target language 0.001009274
output sequence 0.001006946
final accuracy 0.001002777
suitable network 9.9796E-4
same corpus 9.97391E-4
network architectures 9.972029999999999E-4
network pattern 9.95129E-4
network configurations 9.93314E-4
accuracy figures 9.88711E-4
accuracy requirements 9.87617E-4
network hyperparameters 9.87424E-4
