different features 0.00247939
other features 0.002262679
topic segmentation 0.0018259930000000001
local features 0.001798881
linguistic features 0.00178559
prosodic features 0.001666499
sational features 0.001663051
versational features 0.0016617070000000001
different segments 0.0016373730000000001
lexical information 0.001631958
topic segments 0.001610543
cohesion feature 0.001609089
other speech 0.001564117
different learning 0.0015633220000000002
linear topic 0.0015207010000000002
feature analysis 0.001517042
test data 0.001511191
feature tag 0.001505643
topic boundary 0.00150134
size topic 0.001493448
training data 0.0014725060000000002
feature selection 0.001460995
segmentation algorithm 0.001426143
features 0.00142606
topic break 0.0013910230000000001
other test 0.001388491
different lengths 0.001387971
topic seg 0.001384891
topic boundaries 0.001383412
segmentation probability 0.001374344
feature interactions 0.001341294
different levels 0.001336865
different window 0.0013318630000000002
text segmentation 0.001323504
major topic 0.001314863
segmentation approach 0.001309569
little data 0.001308327
topic shifts 0.001307449
classifiers topic 0.00130254
cohesion segmentation 0.001300462
test corpus 0.001298298
other work 0.0012948410000000001
different directions 0.0012879080000000002
lexical cohesion 0.001281445
discourse segmentation 0.001269361
linear model 0.0012639819999999999
topic breaks 0.001262686
many segmentation 0.001261673
labeled topic 0.001261256
tect topic 0.00125906
meeting segmentation 0.001234686
ing corpus 0.001218834
other meetings 0.001201796
other algorithms 0.00119913
speech recognition 0.0011978000000000002
meeting corpus 0.001181619
segmentation problem 0.001180799
cue word 0.00117429
segmentation techniques 0.001165585
segmentation algorithms 0.001162004
segmentation accuracy 0.001154944
conversational speech 0.001136371
information retrieval 0.001134204
ing results 0.001114943
test set 0.001109882
feature 0.00110812
other relations 0.001107039
valuable information 0.001095305
word frequency 0.001084916
speech rate 0.001082305
word distribution 0.001081475
other silences 0.001076415
other segmenters 0.001069761
several parameters 0.001066179
textual segmentation 0.001052334
lexical items 0.001051743
word vector 0.0010489779999999999
lexical chains 0.00103955
lexical chain 0.001034229
agreed segmentation 0.001031842
topic 0.0010265
wsj corpus 0.001026381
word counts 0.001025673
set size 0.0010249579999999999
same test 0.001021661
learning approach 0.001020068
lexical item 0.001019239
content word 0.0010175409999999998
word content 0.0010175409999999998
language models 0.00101446
lexical cohe 0.001014109
speech corpora 0.001012842
stricted model 0.001003519
word usage 9.97892E-4
ing text 9.964190000000001E-4
language research 9.928720000000001E-4
speech activity 9.88795E-4
brown corpus 9.83192E-4
tdt corpus 9.79627E-4
overlapping speech 9.702080000000001E-4
