text segmentation 0.0029684000000000004
segmentation algorithm 0.00296495
same topic 0.002620524
different word 0.0025686069999999997
segmentation corpus 0.0025554370000000002
topic boundaries 0.00248228
same word 0.002480454
probability segmentation 0.002379201
segmentation boundaries 0.00235014
topic sequence 0.002264032
continuous topic 0.0022233590000000003
finding topic 0.002197469
topic transitions 0.002197469
word repetition 0.002170791
segmentation system 0.002160329
word sequence 0.002123962
segmentation accuracy 0.0021186250000000003
segmentation systems 0.002113705
segmentation sys 0.0021111050000000003
word distributions 0.002098695
correct segmentation 0.00209724
story segmentation 0.002087755
likely segmentation 0.0020876650000000003
word tokens 0.0020702579999999997
finer segmentation 0.0020650160000000002
recursive segmentation 0.0020650160000000002
word densities 0.002057705
topic 0.00194935
text segments 0.00188433
different words 0.0018328469999999999
same text 0.001822364
segmentation 0.00181721
text length 0.0016837080000000001
other words 0.001663313
markov model 0.001581542
text seg 0.001571841
lexical cohesion 0.001565224
ing data 0.001533304
statistical model 0.001516782
different topics 0.001495672
long text 0.001472505
training data 0.001466687
original text 0.0014519280000000001
sample text 0.001430532
independent text 0.001411216
text samples 0.001407115
available data 0.0013502660000000001
words increases 0.0013280429999999999
clue words 0.0013268449999999999
data description 0.001326031
stop words 0.0013226199999999998
same segment 0.0012649979999999998
many segments 0.0012515909999999998
various features 0.001240865
test corpus 0.001225749
broadcast news 0.001207182
model 0.00119968
various topics 0.00119535
segment boundary 0.0011627719999999999
algorithm 0.00114774
average length 0.0011412940000000002
segment boundaries 0.001126754
segment length 0.001126342
small segments 0.001114567
tion corpus 0.001092558
words 0.00107352
news story 0.001059565
tation corpus 0.0010540850000000002
information retrieval 0.001042593
average seg 0.001029427
various texts 0.001023405
prior information 0.0010171989999999999
multiple topics 0.001011236
component topics 0.001003737
herent topics 0.001000488
corpus statistics 9.97717E-4
brown corpus 9.918980000000002E-4
artificial corpus 9.90037E-4
ticular topics 9.85964E-4
natural language 9.6733E-4
segment lengths 9.49585E-4
language processing 9.46625E-4
average reference 9.38372E-4
new approach 9.37094E-4
same cost 9.314389999999999E-4
experimental results 9.29855E-4
long document 9.28833E-4
prior probability 9.28694E-4
reference segment 9.234199999999999E-4
large value 9.23297E-4
relative performance 9.00862E-4
small number 8.913199999999999E-4
real texts 8.78696E-4
son distribution 8.73222E-4
average run 8.710549999999999E-4
original results 8.655970000000001E-4
ment texts 8.654369999999999E-4
morphological analysis 8.62486E-4
previous methods 8.618090000000001E-4
corresponding results 8.58503E-4
