word vector 0.002647249
new word 0.0025497609999999998
word vectors 0.0025392789999999998
same words 0.002474528
word utterances 0.0024329
other topic 0.002403314
topic segmentation 0.002350785
original word 0.0023457969999999997
single word 0.0023219
many words 0.00229268
same topic 0.002281038
new topic 0.002262781
unique word 0.00225084
times word 0.0022340899999999998
distinctive words 0.002152992
tween words 0.002146195
individual words 0.0021366420000000002
topic boundaries 0.002130734
topic length 0.002105996
average topic 0.002078461
lexical cohesion 0.002024741
topic shift 0.001990909
topic seg 0.001987717
topic shifts 0.001976708
topic segmenta 0.001946589
topic segmen 0.001925772
mine topic 0.001923663
words 0.00185706
different data 0.001677034
new text 0.0016748610000000001
topic 0.00166357
text unit 0.001653585
lexical approach 0.0016151170000000001
large text 0.001575573
space method 0.0015741980000000002
text units 0.001560376
segment text 0.001528654
text segment 0.001528654
vector space 0.001469671
input text 0.00144841
ment text 0.00143764
space model 0.0014271309999999999
basis cohesion 0.001425772
previous cohesion 0.00137254
text understanding 0.0013711910000000001
new information 0.0013665840000000001
semantic space 0.001353048
cohesion approaches 0.001344531
entire text 0.001339903
test data 0.001338966
dialogue corpus 0.0013337660000000001
different topics 0.001323732
collocational cohesion 0.001309893
lsa method 0.001307049
new vector 0.00129591
basis method 0.001289447
lsa space 0.001278795
dialogue segmentation 0.001273228
vector similarity 0.0012643609999999999
cal cohesion 0.0012500459999999999
current method 0.0012427430000000001
textual information 0.0012347830000000001
lsa algorithm 0.001230192
cohesion gaps 0.001221887
space methods 0.001218495
lsa vector 0.001202522
standard vector 0.001189468
other methods 0.001185267
original space 0.001168219
based method 0.001167489
utterance vector 0.001163785
semantic analysis 0.001156491
vector comparison 0.001147974
different vocabularies 0.00113943
segmentation methods 0.001132738
similar data 0.0011177960000000001
foltz method 0.001116766
data sets 0.001107094
task domain 0.001102907
lsa vectors 0.001094552
method hearst 0.001078089
hearst method 0.001078089
large unit 0.0010778580000000001
standard unit 0.001070704
segmentation criterion 0.001069728
unit size 0.001068674
new utterance 0.001066297
other factors 0.001064382
testing data 0.001061241
information retrieval 0.0010604429999999999
method sensi 0.001058377
foltz algorithm 0.0010399089999999999
linear segmentation 0.001039729
other measures 0.00103811
development corpus 0.001031358
segmentation algorithms 0.001028981
testing corpus 0.001012106
other hand 0.001009541
test utterance 0.001009164
hearst algorithm 0.001001232
