domain data 0.00293364
training data 0.002840917
tagging features 0.002542784
different word 0.002475771
data set 0.002475047
word representation 0.002461319
same word 0.002418887
ing data 0.0024030930000000002
word segmentation 0.0023309150000000002
input data 0.002285412
main data 0.002265825
current word 0.002221284
word embeddings 0.0022046130000000002
word representations 0.002203978
web data 0.00220341
word projection 0.00219608
joint word 0.0021876680000000003
ith word 0.002163786
data sets 0.002146588
only word 0.002140584
unique word 0.002138346
unlabelled data 0.002132162
word combinations 0.00212668
word representa 0.002126556
word types 0.002117678
official data 0.002112452
labelled data 0.0021091449999999998
noisy data 0.00209998
mixed data 0.002099811
word alignment 0.002099332
untagged word 0.002098847
data cleaning 0.002097253
word orders 0.0020880490000000002
word projections 0.002087312
features templates 0.002029364
useful features 0.002009152
indicative features 0.001990422
pos tag 0.001886445
domain tagging 0.001864964
ing feature 0.0018398330000000001
pos tagging 0.001793374
domain sentences 0.001767669
features 0.00175016
feature vectors 0.0017445580000000002
training algorithm 0.0017438829999999999
original feature 0.001644539
standard pos 0.001638178
work learning 0.001623427
feature module 0.0016091200000000001
different training 0.001603558
other pos 0.0015937689999999999
training set 0.0015933639999999999
pos tagger 0.0015864719999999998
feature templates 0.0015772440000000002
model parameters 0.0015753030000000001
feature representa 0.0015727660000000002
supervised learning 0.00155566
model score 0.0015491040000000001
feature repre 0.0015398360000000002
feature fires 0.001534053
target domain 0.00152739
pos tags 0.001513429
statistical model 0.0014839430000000002
learning representations 0.001472398
different words 0.001464568
source domain 0.001450372
unsupervised learning 0.001450009
domain text 0.001443109
learning rate 0.001439258
graphical model 0.0014253830000000001
tagging approach 0.001420431
first pos 0.0014151629999999999
web domain 0.00141445
domain adaptation 0.001402995
divergence learning 0.001395024
tag sequence 0.001383543
test set 0.001381413
learning strategies 0.001357809
domain taggers 0.001345445
unlabelled domain 0.001343202
weblog domain 0.001320912
mixed domain 0.001310851
email domain 0.001309743
feature 0.00129804
training example 0.0012859479999999999
corresponding pos 0.001285511
pos dis 0.001275617
pos taggers 0.001273855
training strategy 0.0012528799999999998
pos disam 0.0012452079999999998
training examples 0.0012441029999999999
pos disambiguation 0.001241437
established pos 0.0012357639999999999
training strategies 0.001217176
tagging input 0.001216736
training encoders 0.001214681
training strate 0.001214681
tagging accuracies 0.001214641
model 0.0011885
such distribution 0.001157833
