features model 0.00369705
model feature 0.00368282
pos features 0.002928163
lexical features 0.002785695
lexical feature 0.002771465
morphological features 0.0027029949999999997
context features 0.00260531
same features 0.002583897
feature selection 0.002498671
feature vector 0.002472521
new features 0.002456958
specific features 0.002435663
text features 0.0024002019999999997
only feature 0.0023864209999999997
feature count 0.002356371
rare features 0.002341334
preliminary feature 0.00233651
certain features 0.002336496
unique feature 0.002311092
ical features 0.002299557
irrelevant features 0.0022976669999999998
dimensional feature 0.002294876
global features 0.0022906099999999998
focused features 0.0022830389999999997
tant features 0.0022830389999999997
feature vec 0.0022733659999999998
feature selec 0.0022724399999999997
svm model 0.002249157
features 0.00208181
classification model 0.0020757279999999998
feature 0.00206758
model type 0.00202199
ing model 0.00199199
new model 0.001990388
model number 0.00197687
model tampering 0.001847036
model tamper 0.001816873
different word 0.001779087
other words 0.001698573
training corpus 0.0016676809999999999
model 0.00161524
current word 0.001603275
word order 0.001562109
test corpus 0.0015237599999999999
corpus error 0.0015221710000000001
corpus errors 0.001507953
english corpus 0.001473051
first words 0.0014648069999999998
free word 0.001458912
rent word 0.00144927
word expressions 0.001437645
word idioms 0.001431251
similar corpus 0.001387876
size corpus 0.001383022
pos errors 0.001322346
corpus variations 0.001296603
hebrew pos 0.001292577
corresponding corpus 0.001267813
hebgold corpus 0.001256755
pos tag 0.001253651
likely corpus 0.001243456
corpus locations 0.001242375
corpus show 0.0012383539999999999
lexical information 0.001238035
different language 0.00123761
erroneous corpus 0.001236278
svm models 0.001234849
suspicious corpus 0.001233933
corpus loca 0.001233933
tified corpus 0.001233933
suspect corpus 0.001233933
pos tags 0.001216221
test data 0.0012122370000000001
classes pos 0.0012002879999999999
specific pos 0.001200206
hebrew data 0.001166661
task data 0.001151826
pos tagset 0.001148651
pos tagging 0.001146486
learning results 0.001106269
pos tagger 0.001096597
svm classification 0.0010944050000000001
wsj pos 0.001076309
training set 0.001072449
new language 0.001063311
syntactic context 0.001059281
words 0.00105398
automatic pos 0.001050856
corpus 0.00103196
other languages 0.001006801
such cases 9.92671E-4
svm analysis 9.90416E-4
learning method 9.82323E-4
svm classifiers 9.71243E-4
other scores 9.65468E-4
other kernel 9.5104E-4
linear svm 9.50698E-4
data points 9.46945E-4
classification errors 9.364810000000001E-4
svm chunking 9.280600000000001E-4
