word model 0.006568289999999999
word segmentation 0.005240595
word boundary 0.005046684
word candidate 0.005026933
candidate word 0.005026933
word sequence 0.005008399
good word 0.004999397
possible word 0.00499636
word sequences 0.0049951299999999995
word problem 0.004986445
stochastic word 0.004933105
word boundaries 0.004928169
frequency word 0.004912270999999999
automatic word 0.0048706949999999995
unknown word 0.004862613
word seg 0.004843913
word segmenter 0.004835615
word spelling 0.00482772
oov word 0.004821542999999999
appropriate word 0.0048164309999999995
proper word 0.004810175
word candidates 0.0048053869999999995
candidates word 0.0048053869999999995
known word 0.004796723
word extraction 0.004791976999999999
word identification 0.0047914459999999996
matic word 0.0047908479999999995
reliable word 0.004789044
usage word 0.004787324999999999
word fragments 0.004787293
obvious word 0.0047861679999999995
word candi 0.0047861679999999995
words model 0.0034405599999999996
language model 0.002798972
segmentation model 0.002747865
model approach 0.0025491249999999997
transcription model 0.0024712149999999997
pronunciation model 0.002398926
phoneme model 0.0023954809999999997
model precision 0.002384775
gram model 0.002373873
guage model 0.002363868
baseline model 0.002360834
channel model 0.0023438449999999998
accurate model 0.002310002
acoustic model 0.0023006529999999997
model the 0.0022959599999999997
nel model 0.002293949
ciation model 0.002293949
model 0.00203778
many words 0.0018986699999999999
new words 0.001841603
content words 0.001751271
unknown words 0.001734883
known words 0.001668993
istic words 0.001658297
target corpus 0.001564284
test corpus 0.001529712
small corpus 0.00150618
domain corpus 0.0014800199999999999
lexical context 0.001455403
corpus size 0.001432043
ing corpus 0.001417755
words 0.00140278
annotated corpus 0.001402118
segmented corpus 0.001395787
raw corpus 0.0013415579999999999
unannotated corpus 0.001299144
corpus usage 0.001297655
character error 0.001193619
same method 0.001190317
segmentation problem 0.0011660199999999998
character sequence 0.001164249
stochastic language 0.0011637870000000001
possible character 0.00115221
character sequences 0.00115098
japanese character 0.001142304
character set 0.001141694
natural language 0.001136288
context information 0.001133662
probability estimation 0.001130872
stochastic segmentation 0.00111268
gram probability 0.001109887
same domain 0.001108612
single character 0.00108618
boundary information 0.001083549
segmentation errors 0.0010675329999999999
first character 0.001059083
probability argmax 0.001051161
automatic segmentation 0.0010502699999999998
segmentation result 0.001044848
corpus 0.00104084
other nlp 0.0010395159999999999
generation probability 0.001038982
probability 日テレ 0.001032774
input method 0.001031389
ation probability 0.001030388
different intonations 0.001029231
same phoneme 0.001027133
other part 0.00102251
