tagging model 0.003237545
pos tag 0.002986766
model baseline 0.002955149
new model 0.002889401
standard model 0.00285349
markov model 0.002853236
extended model 0.002743208
stochastic model 0.002724217
dard model 0.002678058
model infeasible 0.002678058
training data 0.002671679
data set 0.0026367170000000002
ing tag 0.002498293
current tag 0.0024722249999999998
different data 0.002443596
model 0.00242208
test data 0.002401201
corpus data 0.002343155
tag sequences 0.002317507
trigram data 0.00229252
tag trigrams 0.002274158
rent tag 0.002271501
tag filter 0.002268109
data the 0.002137404
data sets 0.00213548
alpino data 0.002078138
wotan data 0.002069115
testing data 0.002060601
extra data 0.002055611
gram data 0.002047505
sparse data 0.002045574
data freq 0.002043099
pos tags 0.001927708
pos tagging 0.001790961
context feature 0.001687564
different tags 0.0016085980000000001
syntactic features 0.001562984
separate feature 0.001512887
same training 0.001508694
possible tags 0.001506063
unknown words 0.001458708
binary feature 0.001445162
tagging method 0.001413179
standard pos 0.0014069059999999999
context information 0.001392943
extra feature 0.001367821
text information 0.001361839
hmm tagging 0.00132328
known words 0.001319279
same probability 0.001311395
baseline accuracy 0.0012821859999999998
new information 0.00127212
hmm tagger 0.001271589
trigram tagger 0.0012690840000000002
initial tags 0.001260062
case information 0.001258877
former features 0.001257008
trigram accuracy 0.001254427
german features 0.001242674
contextual information 0.001237191
preceding tags 0.001223741
chunk tags 0.001215853
probability distribution 0.001212377
syntactical tags 0.0012111090000000001
infrequent tags 0.001210296
bigram accuracy 0.001203786
tagging errors 0.0011802470000000002
order models 0.001172593
syntactic context 0.0011665529999999999
wotan training 0.001166374
initial set 0.001157357
tagging experiments 0.001152894
notated training 0.001152334
other parts 0.001147365
other types 0.001147161
information con 0.001144609
other cases 0.0011423940000000001
possible context 0.0011419949999999998
baseline method 0.0011307829999999998
tagging experiment 0.001128049
same time 0.001126686
frequent tagging 0.001109982
feature 0.00109942
previous context 0.001094912
tagging assumption 0.001088259
extra information 0.0010732
textual information 0.001070982
syntactical information 0.001063696
information incor 0.001061478
new context 0.001055465
current context 0.001049099
words 0.00104531
different sources 0.001043335
large number 0.001033314
tnt tagger 0.001029095
context labels 0.001020883
ging accuracy 0.001015562
trigram hmm 0.0010131250000000001
such error 0.001005357
syntactic relations 0.0010041170000000001
