labeled examples 0.0020199320000000003
unlabeled examples 0.001976654
positive examples 0.001904424
virtual examples 0.0019014790000000002
training data 0.001874239
test examples 0.001865443
examples figure 0.001760005
various examples 0.0017096090000000002
kernel function 0.001652834
examples max 0.0016522330000000001
learning approach 0.001645881
tual examples 0.0016367030000000002
learning methods 0.001634019
decision function 0.001484823
machine learning 0.0014776309999999999
training set 0.00145437
text classification 0.001430377
other method 0.001413812
supervised learning 0.001379764
same label 0.001377082
word fre 0.001372904
examples 0.00136533
active learning 0.00135053
negative example 0.001347718
feature vector 0.0013444899999999998
performance measure 0.001330524
generatebydeletion algorithm 0.0013263480000000002
byaddition algorithm 0.0013263480000000002
deletion algorithm 0.0013263480000000002
binary feature 0.001291111
positive example 0.0012860430000000002
example support 0.001285764
training test 0.001283202
virtual example 0.0012830979999999999
feature vectors 0.0012789049999999999
name training 0.001235631
size training 0.001227284
original example 0.0012029879999999999
effective training 0.0011910940000000002
test set 0.001171394
small training 0.001163501
different methods 0.00115446
same value 0.0011504990000000001
other tasks 0.001137861
original svm 0.0011170120000000001
set size 0.001115476
size set 0.001115476
performance measures 0.001110397
first case 0.001102882
several methods 0.001081868
many features 0.001081731
random number 0.0010807479999999999
virtual support 0.0010749639999999999
training sets 0.001069142
primal performance 0.0010672260000000001
svm figure 0.001055648
imbalanced training 0.001054751
algorithm 0.00105341
performance gain 0.001049843
performance changes 0.001043286
svm framework 0.001040826
combined method 0.001036131
other hand 0.001030295
learning 0.00102747
simple methods 0.0010092599999999999
labeled exam 0.0010001580000000001
other cases 9.9953E-4
experimental results 9.991079999999999E-4
binary classifier 9.95164E-4
set collection 9.85916E-4
function 9.68004E-4
text representation 9.66367E-4
small number 9.63496E-4
ing direction 9.61984E-4
various methods 9.508279999999999E-4
tal results 9.474329999999999E-4
same spirit 9.47342E-4
support vector 9.44263E-4
svm setting 9.41608E-4
nal svm 9.39679E-4
natural language 9.394029999999999E-4
feature 9.39042E-4
svm signifi 9.342930000000001E-4
svm tools 9.342930000000001E-4
combined approach 9.3172E-4
vectors problem 9.172E-4
dard approach 8.95885E-4
target task 8.93161E-4
corpus annota 8.86199E-4
outputs number 8.83511E-4
language processing 8.817949999999999E-4
virtual exam 8.81705E-4
support vectors 8.786779999999999E-4
positive documents 8.699280000000001E-4
decision boundary 8.628209999999999E-4
following steps 8.536700000000001E-4
test exam 8.45669E-4
positive outputs 8.39521E-4
error rate 8.32655E-4
many cases 8.23346E-4
