learning algorithm 0.0024896099999999997
unlabeled examples 0.0023098899999999998
labeled examples 0.002193772
learning method 0.002137364
same examples 0.002113676
learning approach 0.001977875
learning methods 0.001954298
examples figure 0.0019233429999999999
accuracy algorithm 0.0019075949999999998
examples iteration 0.001890878
active learning 0.001882402
training data 0.001876515
selection algorithm 0.001867882
labeled data 0.001822492
sampling algorithm 0.0017909129999999999
beled examples 0.0017726069999999998
learning curve 0.001762896
selected examples 0.001759923
lected examples 0.001759923
learning experiments 0.001747129
supervised learning 0.001737568
unsupervised learning 0.0016906
pool algorithm 0.001679134
tive learning 0.001663796
previous algorithm 0.0016365239999999999
passive learning 0.001630496
japanese word 0.001626849
learning wherew 0.001626687
sive learning 0.001623153
learning itera 0.001623153
ing method 0.0015453849999999998
word boundary 0.001511953
text classification 0.001508204
classification task 0.001483269
good performance 0.001476724
examples 0.00146997
word segmentation 0.001468797
decision function 0.001459264
training set 0.001455626
data sizes 0.001450127
word segmenta 0.0014488769999999999
features accuracy 0.001391673
other words 0.00136696
kernel function 0.001345582
different labeled 0.001343084
learning 0.00133314
new classifier 0.0013089780000000001
previous method 0.001284278
excellent performance 0.001243402
cellent performance 0.001243402
svm framework 0.001234065
current classifier 0.001208154
unlabeled ones 0.0012005549999999999
such methods 0.001196447
image classification 0.001195563
similar results 0.001183349
accuracy curve 0.001180881
large corpus 0.001179856
random sampling 0.001178831
algorithm 0.00115647
japanese corpus 0.0011435360000000001
different pool 0.001141946
experimental results 0.001138345
character set 0.0011313809999999999
test case 0.001121091
probabilistic classifier 0.001119774
final accuracy 0.001108825
classifier increase 0.001107844
corpus annotation 0.001102086
same way 0.0011003649999999999
encoding method 0.001099982
teacher label 0.001088077
tion problem 0.001086319
inal svm 0.0010744679999999999
size pool 0.001072933
pool size 0.001072933
selection algorithms 0.00105742
selection algo 0.001057315
labeled exam 0.001054136
accuracy improvement 0.0010538869999999999
svm tools 0.001049955
same percentage 0.0010491469999999998
accuracy changes 0.0010461709999999998
feature encoding 0.001040389
large pool 0.0010285630000000001
tal results 0.001023473
empirical results 0.001022478
such tasks 0.0010127930000000001
promising approach 0.001004299
first step 9.96877E-4
language process 9.958039999999999E-4
natural language 9.95098E-4
language tasks 9.86813E-4
first attribute 9.80083E-4
different sizes 9.70719E-4
set stan 9.68284E-4
first paper 9.641039999999999E-4
binary features 9.621720000000001E-4
annotation cost 9.58221E-4
general framework 9.561789999999999E-4
