chinese word 0.002081377
word segmentation 0.001732471
word segmenter 0.001693474
word formation 0.001673369
chinese information 0.0013148230000000001
chinese corpus 0.001242753
same distribution 0.001161282
chinese texts 0.0011514379999999999
information theory 0.001036929
english corpus 0.001017272
chinese character 0.00101016
unknown words 9.918050000000001E-4
chinese nlp 9.81178E-4
high value 9.77845E-4
information processing 9.77361E-4
mutual information 9.74364E-4
chinese sentences 9.72653E-4
first step 9.70189E-4
chinese characters 9.36116E-4
chinese corpora 9.313679999999999E-4
ofdts value 9.30346E-4
other types 9.182530000000001E-4
chinese segmenters 9.08088E-4
nmtual information 9.07443E-4
chinese segrnenter 9.040540000000001E-4
distribution ofmi 9.036370000000001E-4
distribution ofdts 8.992690000000001E-4
other ways 8.8865E-4
training corpus 8.88384E-4
new algorithm 8.88107E-4
testing texts 8.83057E-4
distribution graphs 8.806560000000001E-4
tag set 8.80311E-4
such property 8.788940000000001E-4
distribution fmi 8.77081E-4
annotated corpus 8.49857E-4
local min 8.47073E-4
relative measure 8.45017E-4
segmented corpus 8.4241E-4
news corpus 8.42088E-4
raw corpus 8.398209999999999E-4
statistical data 8.264419999999999E-4
new measure 8.23712E-4
input texts 8.18544E-4
local max 7.93E-4
linguistic resources 7.83965E-4
linguistic this 7.83675E-4
first round 7.790519999999999E-4
linguistic resource 7.72556E-4
proper nouns 7.62929E-4
character pairs 7.43339E-4
right side 7.408849999999999E-4
character pair 7.40702E-4
segmentation rules 7.401899999999999E-4
character string 7.36178E-4
input sentence 7.318660000000001E-4
standard eviation 7.29264E-4
local maximum 7.237529999999999E-4
experimental results 7.222330000000001E-4
conditional probability 7.174589999999999E-4
second round 7.15502E-4
absolute measure 7.11794E-4
words 7.03921E-4
data incompleteness 7.02638E-4
local minimum 6.9911E-4
dts values 6.97624E-4
many limitations 6.9747E-4
new concepts 6.97036E-4
additional condition 6.90618E-4
time consuming 6.86402E-4
left side 6.73864E-4
different conditions 6.735389999999999E-4
correct segmentation 6.73082E-4
different domains 6.70274E-4
human supervision 6.663109999999999E-4
main purpose 6.596460000000001E-4
value 6.58703E-4
information 6.58666E-4
human judgment 6.57815E-4
chinese 6.56157E-4
character bigram 6.56053E-4
new domains 6.512110000000001E-4
basic idea 6.5085E-4
construction types 6.49784E-4
natural science 6.473620000000001E-4
nlp system 6.45839E-4
current world 6.45591E-4
various domains 6.43299E-4
input sentences 6.397589999999999E-4
processing system 6.39513E-4
appropriate answer 6.38851E-4
certain degree 6.38751E-4
this work 6.321860000000001E-4
correct france 6.310689999999999E-4
important concepts 6.30234E-4
preliminary experiments 6.287E-4
distribution 6.27626E-4
types ofdts 6.26468E-4
important resources 6.238319999999999E-4
max ifdts 6.23758E-4
