word alignment 0.0029561400000000003
binary feature 0.002668097
feature weights 0.002575539
feature func 0.002464515
feature description 0.002462108
computing feature 0.002440816
word information 0.002298277
lexical features 0.002289936
function word 0.002289915
syntactic features 0.002236781
chinese word 0.002212046
feature 0.00217826
empty node 0.00213118
language model 0.002108284
training data 0.0020881800000000002
head word 0.002038181
word align 0.002037677
label features 0.002037199
tree node 0.002010756
word pairs 0.002009407
original word 0.002004779
supervised word 0.002003783
surface word 0.001980797
incorrect word 0.00197059
frequent word 0.001960734
nese word 0.001960731
features table 0.00195679
prediction model 0.001950105
category model 0.0019398269999999999
automatic word 0.00193484
word aligner 0.001929884
neighboring word 0.001929607
alignment test 0.001918942
annotation model 0.001897869
various features 0.001870656
sparse features 0.001865324
entropy model 0.0018502229999999998
probabilistic model 0.0018434039999999999
similar features 0.0018413869999999999
model scores 0.0018393889999999999
maxent model 0.0018142289999999999
conjunction features 0.0018141799999999999
baseline alignment 0.001745915
parse tree 0.001744214
test data 0.001713902
alignment accuracy 0.001708387
training set 0.001635636
node label 0.001631479
empty nodes 0.0016274459999999998
same training 0.001626088
alignment improvement 0.00161678
syntactic tree 0.0016121170000000002
hand alignment 0.0016029380000000001
incorrect alignment 0.0015933700000000002
system training 0.0015930850000000002
training sentence 0.001593081
child node 0.0015864719999999998
alignment quality 0.001558562
english words 0.001553767
sensible alignment 0.0015494100000000002
alignment avail 0.0015494100000000002
current node 0.00153853
training corpus 0.001534724
model 0.0015339
features 0.00152057
solid node 0.0014948849999999998
parent node 0.001492736
tag lexical 0.001492043
translation grammar 0.0014728520000000002
function words 0.001444406
training time 0.0014435490000000001
ctb data 0.0014345970000000001
empty category 0.001422257
empty elements 0.001413806
tree label 0.001412535
right node 0.0014007639999999999
parse trees 0.001397423
node positions 0.001395045
ancestor node 0.0013947909999999998
ent node 0.0013938239999999999
chinese parse 0.0013936740000000001
real data 0.001389416
rent node 0.001377016
machine translation 0.001375322
chinese words 0.0013665370000000001
empty categories 0.001356929
difficult data 0.001353944
augmented data 0.001353449
tree structure 0.001353117
weblog data 0.001352058
web data 0.0013442060000000001
tree learning 0.001337984
training corpora 0.001335073
machine parse 0.001332213
parallel training 0.001314655
translation task 0.001290584
alignment 0.00128946
such trees 0.001286047
empty detection 0.001275774
target words 0.0012729080000000001
