crf model 0.00285889
feature set 0.002602725
standard feature 0.002325056
baseline feature 0.002311222
first feature 0.0022967179999999997
transition feature 0.0022913119999999998
model accuracy 0.0022317960000000003
feature functions 0.002193731
feature sets 0.002189425
model learning 0.002186532
training data 0.00216943
feature extraction 0.002119312
feature expansion 0.002098973
feature marker 0.0020892669999999997
emission feature 0.002086975
straightforward feature 0.002084823
feature markers 0.002084644
markov model 0.002083433
feature functionsx 0.002080072
model estimation 0.002049535
chains model 0.002033105
model parameters 0.002003435
model gender 0.00197905
model order 0.001930202
model specification 0.001908094
label set 0.001899595
standard features 0.0018623860000000002
feature 0.00185284
training set 0.001845695
transition features 0.0018286420000000001
data set 0.001823505
same data 0.001754936
word form 0.0017288330000000001
model 0.00168143
tion features 0.001639222
emission features 0.001624305
ing crf 0.001612114
word position 0.0016114410000000001
label transition 0.001588182
word forms 0.001585388
word class 0.001543681
full label 0.001542002
known word 0.001536061
english crf 0.001525504
nominative word 0.0014970810000000001
label sets 0.0014862949999999999
label transitions 0.001445011
romanian crf 0.001442339
finnish crf 0.00141168
corresponding label 0.001410434
data sets 0.0014102049999999999
estonian crf 0.001404422
czech crf 0.001404422
pos labels 0.001397289
features 0.00139017
label partitions 0.0013860629999999999
pound label 0.001380048
compound label 0.0013799349999999999
coarse label 0.001379304
ple label 0.001376745
test set 0.001375416
third data 0.001362388
training times 0.001353861
data sparsity 0.001339585
east data 0.001318909
data spar 0.001300236
tag set 0.00125012
pos sequence 0.001241934
tagging approach 0.001241498
other words 0.001189296
label 0.00114971
pos tagging 0.001149381
training 0.00109581
test tags 0.0010882420000000001
conditional random 0.001085148
annotation approach 0.001073533
extraction approach 0.0010330180000000001
tagging accuracy 0.001025318
markov models 0.001023672
development set 0.001022076
preferable approach 9.96492E-4
tionary approach 9.96492E-4
set sizes 9.86537E-4
adjacent labels 9.81648E-4
pos annotation 9.814160000000001E-4
unordered set 9.76218E-4
chains models 9.733440000000001E-4
pos tagger 9.670410000000001E-4
test sets 9.621160000000001E-4
compound labels 9.53085E-4
perceptron algorithm 9.308019999999999E-4
morphological tag 9.266980000000001E-4
error analysis 9.24097E-4
different annotation 9.16689E-4
hunpos results 9.07754E-4
dependency treebank 9.045889999999999E-4
complete test 8.95615E-4
several problems 8.948529999999999E-4
sentence position 8.8233E-4
penn treebank 8.80371E-4
