same model 0.002789474
model parameters 0.002566981
model structure 0.002383174
generative model 0.002291472
markov model 0.002283571
same tag 0.0022747740000000002
model definition 0.0022443429999999998
field model 0.002235065
our model 0.002201624
model vari 0.002168016
discriminative model 0.002161348
native model 0.002159678
different tag 0.002116702
same word 0.0020559339999999997
other tag 0.002040903
tag dictionary 0.001983057
tag information 0.0019479060000000001
model 0.00190304
random tag 0.001896573
tag accuracy 0.001887128
same data 0.00185397
large tag 0.0018324320000000002
full tag 0.001775819
tag class 0.0017662580000000002
tag sequence 0.001747943
tag assignment 0.001718595
map tag 0.001703273
function words 0.0017023160000000002
trigram tag 0.001674774
tag assignments 0.0016682090000000001
ith tag 0.001655659
tag trigrams 0.001646151
guished tag 0.00164473
tag dictio 0.00164473
pos tags 0.0016336800000000002
pos tagging 0.0015988920000000002
sampling distribution 0.0015967759999999998
data set 0.0015906499999999999
other words 0.001576331
dirichlet distribution 0.001570608
posterior distribution 0.001557669
prior distribution 0.0015516969999999999
training corpus 0.001549265
unsupervised pos 0.001526428
word classes 0.001523271
word types 0.001512929
same set 0.0015095479999999999
uniform distribution 0.001488156
possible words 0.0014798950000000002
word type 0.001461257
same number 0.001427565
multinomial distribution 0.00139978
corpus size 0.001396117
normal distribution 0.001389282
standard results 0.001386073
same accuracy 0.001385222
conditional distribution 0.001381312
bayesian approach 0.001366487
output distribution 0.001356231
proposal distribution 0.001352512
tional distribution 0.001339047
wsj corpus 0.0013360989999999999
same form 0.001331295
transition probability 0.001325696
language learning 0.001318437
other tags 0.001314549
entire data 0.001292066
unlabeled data 0.001290167
data points 0.00128769
same cluster 0.001282618
prt pos 0.001280007
bayesian hmm 0.0012722010000000001
pos lpunc 0.001268098
data sets 0.001254296
pos disambiguation 0.001253647
corpus sizes 0.001222955
tagging dictionary 0.001221915
standard tags 0.001220942
infinite models 0.001220928
different tokens 0.0012206869999999998
possible parameters 0.001220068
possible tags 0.001218113
unsupervised learning 0.001200242
linguistic models 0.001199383
infrequent words 0.001195967
various models 0.001192934
corpus stats 0.001190012
possible parameter 0.0011895719999999999
varying corpus 0.0011891480000000001
high probability 0.001183033
content words 0.001180503
optimal results 0.001175113
dictionary information 0.0011542829999999999
equal probability 0.001152531
bhmm results 0.0011519920000000001
hmm structure 0.001143129
different corpora 0.001139596
probability matrix 0.0011384350000000001
tagging accuracy 0.001125986
parameter values 0.001117825
