word features 0.00274527
training data 0.002218204
twitter data 0.002216755
word type 0.002030524
ing data 0.002006523
test data 0.001973158
word con 0.001970343
labeled data 0.0019528050000000002
english data 0.001944514
middle word 0.0019141429999999999
word shapes 0.001911851
annotated data 0.001897859
media data 0.0018905200000000001
data sets 0.001806365
development data 0.00177135
annotate data 0.001757879
pos tagging 0.001674446
pos tagger 0.001630195
twitter pos 0.001502606
other twitter 0.001490899
distributional features 0.001439333
feature set 0.0014019129999999999
local features 0.001383637
annotation annotation 0.00135464
base features 0.001339832
certain features 0.001339569
ptb words 0.001330066
foreign words 0.0013122490000000001
partial words 0.001310998
tag type 0.001291017
feature vector 0.001279889
second feature 0.0012555679999999999
tag description 0.001252822
tag dictionaries 0.001245542
english pos 0.001230365
machine learning 0.001217954
full model 0.0012162240000000001
tagging results 0.00121539
full feature 0.001214482
tag dictionary 0.001196116
ptb tag 0.0011915390000000001
traditional tag 0.001185773
frequent tag 0.00118405
final tagging 0.001182782
confused tag 0.001181601
pos tagset 0.001181471
tag acc 0.001179257
pos taggers 0.001172724
stanford pos 0.001171552
stanford tagger 0.0011694449999999999
base model 0.001164165
feature ablation 0.0011487490000000001
pos categories 0.001148107
tagging scheme 0.0011456650000000001
difficult tagging 0.001138502
other categories 0.0011364
names feature 0.001115733
new tags 0.001111835
tagger improvements 0.001107479
tagging consistency 0.001098148
tagging guidelines 0.0010936890000000001
report tagging 0.0010909000000000001
tagging deci 0.0010860470000000001
words 0.0010824
linguistic analysis 0.001082354
our tagger 0.0010815059999999999
other abbreviations 0.001070355
features 0.00106189
traditional pos 0.001058051
trained pos 0.001056927
pos inventory 0.00104356
correct tags 0.001037673
twitter discourse 0.001031771
nlp system 0.001030372
twitter users 0.0010201659999999999
twitter categories 0.0010184109999999999
set accuracy 0.0010168529999999999
english corpus 0.0010091850000000001
stanford tags 0.001006417
text analysis 9.76368E-4
special twitter 9.69526E-4
manual annotation 9.69349E-4
annotation scheme 9.6469E-4
annotated corpus 9.625300000000001E-4
labeled text 9.61483E-4
test set 9.602899999999999E-4
standard set 9.56108E-4
english text 9.53192E-4
twitter orthography 9.52422E-4
annotation process 9.49426E-4
language technology 9.49181E-4
different tagsets 9.46655E-4
linguistic resources 9.460860000000001E-4
test accuracy 9.422790000000001E-4
stanford system 9.35747E-4
learning 9.29886E-4
ual annotation 9.28299E-4
service twitter 9.241119999999999E-4
case study 9.19329E-4
same time 9.19199E-4
