answer questions 0.00267215
language model 0.002314825
feature set 0.002279872
candidate answer 0.002148512
answer class 0.002093792
word answers 0.002088424
correct answer 0.002076109
answer sequence 0.002055473
answer accuracy 0.002051706
optimal answer 0.002012086
answer typing 0.001982121
answer candidates 0.001980315
answer classes 0.001977134
retrieval model 0.001952082
statistical model 0.001892042
above model 0.001811235
trieval model 0.001775057
guage model 0.001742183
filter model 0.001740859
ter model 0.001740206
answer 0.00169765
other words 0.001581748
feature 0.00155454
language models 0.001492598
evaluation set 0.001486648
single word 0.001482337
other information 0.0014764140000000001
model 0.00147038
input question 0.001458891
question type 0.001456314
question classification 0.001445517
question answering 0.001432624
word modeling 0.00142846
example questions 0.001403359
example answers 0.001395153
predefined question 0.00139102
question categories 0.001387866
data set 0.001378851
type features 0.001375884
training data 0.0013749629999999999
training pairs 0.001369638
baseline system 0.001362155
same data 0.0013540639999999999
other cluster 0.001309379
same cluster 0.001246343
class language 0.001240587
accurate answers 0.0012300380000000001
ble answers 0.0012300380000000001
natural language 0.001200421
information retrieval 0.001194535
training pair 0.001186358
evaluation metric 0.001183497
clustering algorithm 0.001171294
frequency words 0.001152744
frequent words 0.0011456399999999999
language mod 0.0011297009999999999
few words 0.001120759
question 0.00110769
development set 0.001094909
algorithm iterations 0.001053418
other languages 0.001038794
indicator function 0.001035238
features 0.00102726
joint set 0.001022837
ing pairs 0.001003678
greedy algorithm 9.91303E-4
mapping function 9.882699999999999E-4
ery training 9.83519E-4
data sets 9.83371E-4
questions 9.745E-4
answers 9.66294E-4
age models 9.54351E-4
learning methods 9.444289999999999E-4
data points 9.33652E-4
correct probability 9.335120000000001E-4
cluster configuration 9.24355E-4
several trec 9.198749999999999E-4
data sparsity 9.17633E-4
probability distribution 9.12569E-4
speech recognition 8.77006E-4
system 8.75469E-4
fixed cluster 8.71583E-4
experimental results 8.665719999999999E-4
original cluster 8.63433E-4
retrieval tasks 8.4947E-4
language 8.44445E-4
optimal number 8.421750000000001E-4
clustering procedure 8.347529999999999E-4
own cluster 8.34148E-4
correct candidate 8.29321E-4
aquaint corpus 8.28657E-4
cluster configura 8.27616E-4
cluster end 8.25899E-4
cluster configurations 8.25823E-4
support vector 8.25707E-4
words 8.18167E-4
cluster size 8.151149999999999E-4
manual clusters 8.11318E-4
cbest cluster 8.11084E-4
empty cluster 8.11084E-4
