other word 0.00324466
word embeddings 0.003008932
use word 0.003005417
word vectors 0.002977389
context word 0.002943729
text word 0.002915651
word distribution 0.002900319
word distributions 0.002891677
single word 0.002852123
word represen 0.002828831
linear model 0.0028127100000000004
word surfaces 0.002804188
word distri 0.002803093
pos tag 0.00275818
same features 0.0025923480000000004
tagging model 0.002541354
distributional features 0.0025103010000000004
tion features 0.002491737
discrete features 0.0024771150000000002
continuous features 0.002461674
binary features 0.002444031
distribution features 0.0024439690000000003
raw features 0.0023920250000000003
supertag features 0.0023808650000000002
combinatorial features 0.002378495
dense features 0.0023657070000000003
features com 0.002361757
tinuous features 0.00235066
uous features 0.002348577
discretized features 0.002347113
crete features 0.002347113
pos tagging 0.002341604
joint model 0.002262236
model parameters 0.002133538
pos tags 0.002128196
features 0.00210835
pos tagger 0.002103464
pos accuracy 0.002048084
hybrid model 0.0020306630000000003
same pos 0.002024458
model combinations 0.002021275
markov model 0.002017327
previous pos 0.002015785
ear model 0.001998674
model flinear 0.001991389
pos accuracies 0.00198247
ensemble model 0.00197898
training data 0.0019391909999999998
corresponding pos 0.001910814
gold pos 0.001895825
english pos 0.001890695
pos predictions 0.0018401770000000001
art pos 0.001822776
special feature 0.0018208349999999998
feature space 0.00181818
tag sequence 0.00178289
nese pos 0.00178213
feature representation 0.001778916
feature vectors 0.0017686989999999999
feature values 0.001760233
function function 0.001753746
model 0.00174021
corresponding feature 0.001726364
continuous feature 0.0017093339999999999
linear models 0.0017084350000000001
binary feature 0.001691691
distribution feature 0.001691629
feature engineering 0.001654528
feature counts 0.001646397
other words 0.001632704
arbitrary feature 0.0016288119999999999
feature mapping 0.0016214889999999998
deterministic tag 0.001613295
feature entries 0.001594723
training corpus 0.0015888080000000001
data set 0.0015754279999999998
tag distribution 0.001553339
tag distributions 0.001544697
same data 0.001499828
training set 0.001482959
ing data 0.0014750289999999999
tag dis 0.001457538
algorithm function 0.001447928
function algorithm 0.001447928
linear part 0.001415255
linear interactions 0.001413807
development data 0.001410574
linear units 0.0014042820000000002
tagging tasks 0.001391542
tagging algorithm 0.0013721990000000002
feature 0.00135601
ing function 0.001336072
linear com 0.001325907
linear classifiers 0.0013200570000000001
linear combination 0.001318198
loss function 0.001305188
data plots 0.00129318
data pairs 0.0012851859999999998
data sets 0.00127704
ptb data 0.0012560199999999998
