word relations 0.002759984
english word 0.002701507
important word 0.002667638
word types 0.002631718
query language 0.002621636
word relationships 0.002611886
similar word 0.002608313
possible word 0.002582218
pairwise word 0.002553917
use word 0.002517209
informative word 0.002507606
word combinations 0.002492257
trivial word 0.002482543
word tokens 0.002481658
conjunctive word 0.002475187
noisy word 0.002475187
search query 0.002470696
same query 0.002407853
semantic features 0.002221054
base query 0.002213614
query log 0.002206145
query likelihood 0.002158087
query representation 0.002139756
query noun 0.002138847
semantic information 0.00211856
query bigrams 0.002089276
query remainder 0.002069814
unigram query 0.002067348
lated query 0.002065565
query format 0.002065318
single words 0.002008732
related words 0.0019516029999999999
content words 0.0018764469999999998
catena words 0.001865849
individual words 0.001859252
elided words 0.001844715
query 0.00184273
component words 0.001837783
stop words 0.001833603
semantic relations 0.001819134
language model 0.0017381879999999999
translation model 0.001654782
semantic role 0.001642841
semantic units 0.001616971
words 0.00159951
semantic relationship 0.001595028
semantic content 0.001589057
semantic classification 0.001585239
semantic unit 0.0015396869999999999
semantic roles 0.001535366
semantic properties 0.001534315
semantic kernels 0.001533631
information retrieval 0.00150715
language models 0.001428613
several features 0.001380318
candidate features 0.001336395
path features 0.00130386
same queries 0.001268363
ellipsis features 0.001257311
linguistic features 0.001254064
syntactic language 0.001239201
prediction features 0.001229845
text processing 0.0011988580000000001
dependency tree 0.001193872
classifier features 0.001193086
feature type 0.001186378
classification features 0.001182053
heuristic features 0.001168563
weak features 0.00116417
short text 0.001157525
sentence similarity 0.001150744
surface text 0.001141145
pos language 0.001138852
characteristic features 0.001130876
ngram features 0.001130876
information gain 0.0011218629999999999
language processing 0.001118992
dependency relations 0.001106808
elided text 0.0011039770000000001
text remnant 0.001087876
text segments 0.001082367
important feature 0.0010818989999999999
text seg 0.001081627
weighted information 0.001076567
dependency graph 0.0010400449999999999
document frequency 0.001036975
retrieval collection 0.001036878
average score 0.001028264
explicit information 0.001028242
single dependency 0.001009016
training data 0.0010063429999999998
search terms 0.001005103
automated language 0.001002408
second feature 0.001001549
dependency path 9.9472E-4
machine translation 9.93236E-4
other node 9.9225E-4
boolean feature 9.64931E-4
dependency paths 9.625629999999999E-4
model 9.59282E-4
