word sets 0.003465623
word window 0.003436861
ith word 0.003402987
sentence segmentation 0.002631413
segmentation position 0.002265136
segmentation rules 0.0021870220000000003
segmentation error 0.00218543
segmentation positions 0.002179443
segmentation performance 0.002166833
segmentation accuracy 0.002160786
segmentation method 0.002143117
segmentation approaches 0.002120964
segmentation improvement 0.002119761
segmentation target 0.0021151250000000003
explicit segmentation 0.002104689
ions segmentation 0.002078277
potential segmentation 0.002074956
appropriate segmentation 0.002072239
segmentation coverage 0.002067378
safe segmentation 0.002062285
mark segmentation 0.0020576830000000003
segmentation posi 0.002052865
segmentation appropriateness 0.002048307
unsafe segmentation 0.00204717
mining segmentation 0.002044896
segmentation tar 0.002044896
segmentation ratio 0.002044896
ungafe segmentation 0.002044896
segmentation cover 0.002044896
entropy model 0.002031751
exponential model 0.001999257
model construction 0.0019187879999999998
tropy model 0.0019173559999999998
trainable model 0.0019147189999999998
segmentation 0.00181462
training data 0.0017291680000000001
model 0.00168419
consistent words 0.0016333239999999998
probability distribution 0.001597837
marked words 0.0015794259999999999
text corpus 0.0014715000000000002
words 0.00134831
simple sentence 0.001339916
translation system 0.00130363
machine translation 0.001295028
sentence parsing 0.001283558
feature set 0.001247323
ing sentence 0.001238235
unseen data 0.001225238
sentence boundary 0.001216638
sentence results 0.00121017
training time 0.001188501
sentence analysis 0.001173949
accurate translation 0.001160662
translation systems 0.001154221
coordinate sentence 0.001130294
chine translation 0.001125415
translation archive 0.001123626
target sentence 0.001117298
complex sentence 0.001116159
long sentence 0.001108973
sentence structure 0.001097901
corpus anno 0.001081311
sentence segmentat 0.001075024
empirical distribution 0.001073585
sentence patterns 0.001059384
global sentence 0.001057314
only training 0.001055128
sentence isnot 0.001050306
training portion 0.001048455
ing algorithm 0.001044054
much training 0.001037892
lexical context 0.00103608
contextual information 0.001029551
entropy distribution 0.001022953
test sentences 0.001013138
lexical con 0.001008841
training sample 0.001008001
simple sentences 0.001006468
candidate features 0.0010006149999999998
context information 9.86848E-4
similar features 9.66951E-4
ing test 9.512349999999999E-4
english sentences 9.47623E-4
statistical models 9.30153E-4
uniform distribution 9.2746E-4
probability 9.22445E-4
ability distribution 9.18472E-4
distribution consis 9.07976E-4
ion schemes 9.03037E-4
lexical contexts 9.02012E-4
lexical contextua 8.967609999999999E-4
translation 8.92709E-4
lexical contex 8.88785E-4
lus ion 8.81062E-4
such sentences 8.688459999999999E-4
active features 8.658699999999999E-4
segmentat ion 8.62308E-4
text categorization 8.55508E-4
error sentences 8.54155E-4
