feature model 0.004318799999999999
lexical model 0.003199861
lexical features 0.002964871
svm model 0.002870489
model performance 0.002869996
other model 0.002822567
model accuracy 0.002776588
such features 0.002726055
such feature 0.0027074449999999997
new model 0.002704833
last model 0.002625136
parsing model 0.002590002
model analysis 0.0025682359999999998
large feature 0.0025444889999999996
chored model 0.002489373
model suf 0.002488142
possible features 0.002443926
feature space 0.0024390519999999997
feature selection 0.002431141
pruning features 0.0024304039999999997
following features 0.002420543
feature pruning 0.002411794
binary feature 0.0023576839999999996
specific features 0.00233995
available features 0.002329821
feature pairs 0.002311293
unique feature 0.0023045329999999997
feature sets 0.0023019119999999997
rare features 0.002296378
robust features 0.002296267
model 0.0022862
tive features 0.002286143
cal features 0.002280333
active features 0.002278329
feature representation 0.0022701189999999997
common feature 0.00226679
only features 0.002266085
distinct features 0.002263308
feature vectors 0.0022628229999999997
suffix features 0.002259502
bilexical features 0.002255606
chor features 0.002253285
anchor features 0.002253285
sive feature 0.0022528739999999998
proposed feature 0.00224212
feature spaces 0.002237997
ort feature 0.002235154
canbeprep feature 0.002235154
feature representa 0.002235154
cessful feature 0.002235154
training corpus 0.002100919
features 0.00205121
feature 0.0020326
training data 0.0019932450000000003
lexical models 0.001794988
language data 0.0017055479999999999
function words 0.0016030509999999999
training text 0.0016030320000000001
training set 0.001599013
lexical information 0.001485915
word form 0.001472789
svm models 0.0014656159999999999
text corpus 0.001463571
main word 0.001453377
training corpora 0.0014390260000000002
different results 0.0014188590000000002
word forms 0.001411896
exact word 0.0014115019999999998
other data 0.001409422
training example 0.001390116
test data 0.001359897
training material 0.001332499
training instances 0.001323153
canonic training 0.0013220760000000002
complete word 0.001316166
idiomatic word 0.0013160469999999999
lexical fea 0.001313952
word usage 0.001299581
linear models 0.001277568
corpus source 0.0012760900000000001
current models 0.001274645
word oper 0.0012667969999999998
wsj corpus 0.001262682
lexical forms 0.0012607669999999999
neighbouring words 0.0012488199999999999
surrounding words 0.001247289
different text 0.001245115
such function 0.001233706
genres corpus 0.0012260880000000002
annotated corpus 0.0012110480000000002
standard models 0.001203862
testing corpus 0.0012003860000000001
lexical counts 0.001196528
corpus com 0.0011950790000000001
nlp models 0.001194207
cific corpus 0.001187043
corpus variation 0.001185187
notated corpus 0.001183362
lexical items 0.001177118
data point 0.001163246
