word segmentation 0.00467536
chinese segmentation 0.00357522
different segmentation 0.0035106950000000003
segmentation algorithm 0.003441968
initial segmentation 0.003214006
word model 0.0031659899999999996
chinese word 0.0031542
english segmentation 0.003140136
segmentation results 0.0030674020000000003
segmentation work 0.003058641
segmentation algorithms 0.003056082
many segmentation 0.003035668
high segmentation 0.003023764
segmentation output 0.002994709
segmentation score 0.002925097
segmentation scores 0.002912249
several segmentation 0.00290739
native segmentation 0.002895846
segmentation schemes 0.002888608
segmentation accuracy 0.002880931
segmentation algo 0.002872294
accurate segmentation 0.002868518
match segmentation 0.002867335
segmentation transformations 0.002866634
ent segmentation 0.002853576
low segmentation 0.002852962
segmentation approximation 0.002850065
segmentation experiments 0.002838474
segmentation transfor 0.002838111
nese segmentation 0.002825381
segmentation decisions 0.002823804
tial segmentation 0.002822585
glish segmentation 0.002820991
segmentation papers 0.002820951
segmentation module 0.002819517
predetermined segmentation 0.002819517
word length 0.0027240619999999998
english word 0.0027191159999999997
word seg 0.002615625
average word 0.002590219
segmentation 0.00254819
word boundaries 0.0025107439999999996
form word 0.002507658
word segmen 0.0025070049999999997
single word 0.0025033919999999997
thai word 0.0024897699999999997
chinese words 0.0024716
word list 0.0024698159999999997
accurate word 0.0024474979999999998
word separation 0.0024423649999999997
separate word 0.0024195849999999997
global word 0.002418249
word lists 0.0024146669999999997
distinct word 0.002409665
training data 0.00223279
chinese data 0.0020846700000000003
english words 0.002036516
chinese information 0.0019084549999999999
training sentences 0.001844203
training set 0.001830255
thai words 0.00180717
english training 0.0017670960000000001
words com 0.001762918
test data 0.001753864
full words 0.0017311190000000001
sentence training 0.0017180210000000001
enough words 0.0017180110000000002
initial model 0.0017046359999999998
character set 0.001683725
english data 0.0016495860000000002
character sequence 0.0016242840000000001
ing data 0.0015739080000000002
chinese performance 0.001562888
initial algorithm 0.001559594
same corpus 0.0015561770000000002
chinese results 0.001546242
ing chinese 0.0015432979999999998
chinese segmenter 0.0015081229999999999
sequence algorithm 0.0014894420000000001
first character 0.001477753
different algorithms 0.001470397
appropriate training 0.001467115
entire training 0.001458353
different seg 0.00145096
uring training 0.001446718
words 0.00144457
chinese texts 0.001443384
algorithm performance 0.001429636
next character 0.001423971
thai data 0.0014202400000000001
learning algorithm 0.001419838
single character 0.0014048420000000001
large corpus 0.001388555
size chinese 0.001385889
thai corpus 0.0013672600000000001
simple algorithm 0.0013548499999999999
test set 0.001351329
chinese person 0.0013290239999999998
chinese experiments 0.001317314
same text 0.001307197
