parser training 0.0031847
training algorithm 0.002785193
training data 0.002784042
training time 0.002491851
training parse 0.002476066
feature set 0.002418203
training parsers 0.002399252
discriminative training 0.002324233
baseline training 0.00231885
feature vector 0.002297685
parsing algorithm 0.002248683
training method 0.00224066
single training 0.002203958
large feature 0.002175942
training examples 0.002165967
training procedure 0.002153568
training iteration 0.002129788
training inference 0.002125046
various feature 0.002105976
training speed 0.002101775
our training 0.002101656
training inferences 0.002090605
training run 0.002090182
training process 0.002081141
training times 0.002074452
training regime 0.002047218
stop training 0.002047218
compound feature 0.002015331
splitting feature 0.002007214
feature sets 0.001998177
atomic feature 0.001997304
feature space 0.001996595
feature selection 0.001960267
exponential feature 0.0019383949999999999
ponential feature 0.001932322
pound feature 0.001932322
feature cutoffs 0.001932322
tal feature 0.001932322
feature engi 0.001932322
different parsing 0.001931855
deterministic parser 0.00191237
discriminative parser 0.001910033
training 0.00179945
discriminative parsing 0.001787723
parsing method 0.00170415
sink parser 0.001690515
feature 0.00168397
perceptron algorithm 0.0016661000000000002
word sentences 0.0016524859999999999
learning algorithm 0.001648008
parser searches 0.001639966
working parser 0.001634823
parsing strategy 0.001613455
loss function 0.0015873369999999999
parse state 0.00158039
tive parsing 0.00156045
generative model 0.001545773
constituent parsing 0.001539993
parsing strategies 0.001526033
stituent parsing 0.001525991
shallow parsing 0.0015127119999999998
test data 0.00149694
current model 0.001468816
linear model 0.001457038
parse tree 0.0014398290000000001
collins model 0.0014251770000000001
word sen 0.0014205469999999999
tive model 0.00141877
model selection 0.001397557
parser 0.00138525
ing data 0.001347894
length features 0.001345398
correct state 0.001303657
compound features 0.0012964860000000001
punctuation features 0.001293354
context features 0.001292098
atomic features 0.001278459
arbitrary features 0.0012752190000000002
child item 0.00127517
different parsers 0.001268717
ing state 0.001267076
parsing 0.00126294
head item 0.0012567569999999998
baseline set 0.001253633
parameter vector 0.0012485690000000002
test set 0.001246581
algorithm the 0.0012400170000000002
irrelevant features 0.001234407
active features 0.001230935
sophisticated features 0.0012244080000000002
input sentence 0.001220259
decision tree 0.001212929
discriminative learning 0.001187048
other parsers 0.0011798310000000001
initial state 0.001177907
loss gradients 0.001165858
unpenalized loss 0.001164899
successor state 0.001152869
complete parse 0.0011509229999999999
discriminative parsers 0.001124585
