word segmentation 0.0036847900000000003
model models 0.002679969
unigram model 0.0026381200000000003
bigram model 0.0026339180000000003
model description 0.002402668
gram model 0.0023897320000000003
word boundaries 0.002366157
actual word 0.00228666
word dependencies 0.002285976
word tokens 0.002281626
word type 0.002279049
individual word 0.002260681
segmentation problem 0.00224744
word sequences 0.002233368
ble word 0.002182515
segmentation performance 0.0021814580000000003
model 0.00211616
segmentation scores 0.002109264
current segmentation 0.002101326
specific segmentation 0.002075087
proposal segmentation 0.002006166
original segmentation 0.0019974290000000002
segmentation literature 0.0019862070000000003
quality segmentation 0.00197533
segmentation 0.00173469
adjacent words 0.0015648349999999999
inference algorithm 0.001351751
words 0.00133052
conditional distribution 0.001315211
sampling algorithm 0.00130112
learning algorithm 0.001290115
base distribution 0.001240568
posterior distribution 0.001237325
training data 0.001218733
uniform distribution 0.001217965
proposal distribution 0.001200495
terior distribution 0.001163587
dictive distribution 0.001156483
algorithm samples 0.001121366
incremental algorithm 0.001117091
mcmc algorithm 0.001114032
programming algorithm 0.0011055919999999999
filter algorithm 0.001100011
high probability 0.0010732229999999999
inference problem 0.001052785
data increases 0.001051088
unigram bigram 0.001039718
bayesian inference 0.001036335
different proposal 0.0010345929999999999
probability solutions 0.0010186729999999999
probability particles 9.96366E-4
posterior probability 9.90789E-4
inference performance 9.86803E-4
previous results 9.859259999999998E-4
language acquisition 9.83104E-4
chinese restaurant 9.6962E-4
large number 9.62889E-4
previous work 9.364099999999999E-4
distribution 9.29019E-4
models dependencies 8.996849999999999E-4
inference methods 8.874239999999999E-4
small number 8.71625E-4
speech utterances 8.65165E-4
inference algorithms 8.46354E-4
only inference 8.44497E-4
probabilistic inference 8.38877E-4
same parent 8.356349999999999E-4
batch sampling 8.33187E-4
good performance 8.15201E-4
algorithm 8.11716E-4
new table 8.09396E-4
current state 8.08459E-4
grammatical inference 8.040359999999999E-4
unsupervised way 8.02048E-4
corpus 8.01728E-4
gibbs sampler 8.013779999999999E-4
sampling algorithms 7.95723E-4
learning algorithms 7.84718E-4
incremental learning 7.83774E-4
programming sampling 7.8328E-4
standard batch 7.810390000000001E-4
approximate inference 7.80237E-4
further research 7.641729999999999E-4
ture research 7.621489999999999E-4
segmented utterance 7.62079E-4
minor error 7.58389E-4
time step 7.38885E-4
generative process 7.36602E-4
online learning 7.34329E-4
language 7.33795E-4
many tables 7.26196E-4
ing rejuvenation 7.1426E-4
possible segmentations 7.13974E-4
segmented string 7.11597E-4
batch sampler 7.09235E-4
learning constraint 7.08996E-4
fixed number 7.06777E-4
ticle number 7.05448E-4
state space 6.93817E-4
batch markov 6.84452E-4
