training data 0.0032559999999999998
data quality 0.0029118879999999996
data set 0.002772111
annotated data 0.002688035
ing data 0.0026519559999999996
noisy data 0.002549529
data points 0.0024976
data sets 0.0024814499999999996
pool data 0.00247361
initial data 0.002466162
seed data 0.002464199
real data 0.0024344519999999997
unannotated data 0.0024249469999999998
erroneous data 0.0024042499999999997
tated data 0.0024041419999999997
sufficient data 0.002403381
goldstandard data 0.0024032719999999997
artificial data 0.0024030139999999998
dard data 0.0024017129999999998
annotation noise 0.002305803
human annotation 0.002239276
different word 0.002049225
learning classifier 0.001988862
annotation process 0.001951637
annotation decision 0.0019389499999999998
man annotation 0.001883035
annotation decisions 0.001860556
word sense 0.001837745
annotation schemes 0.0018315929999999998
annotation scenario 0.0018166889999999998
blind annotation 0.001816065
different sampling 0.001811297
annotation scheme 0.001811187
inconsistent annotation 0.0018067839999999999
systematic annotation 0.0018033049999999998
learning task 0.00180246
annotation judgements 0.001796701
active learning 0.001786059
training set 0.001777131
different labels 0.001725971
machine learning 0.001703505
learning process 0.001681287
sampling methods 0.0016608109999999999
training size 0.001659743
learning techniques 0.001623905
random sampling 0.001583181
tive learning 0.001581191
word senses 0.001532365
learning trial 0.001525345
sampling method 0.001523275
annotation 0.00151992
classifier performance 0.001458005
random noise 0.0014549419999999999
uncertainty sampling 0.0013674009999999999
human annotations 0.001311633
entropy sampling 0.001307659
standard sampling 0.00130076
text classification 0.00129581
different types 0.0012637199999999999
learning 0.00124957
different levels 0.001247134
sampling baseline 0.0012439859999999999
noise uncertainty 0.001239162
margin sampling 0.001230424
different parameters 0.00121116
different settings 0.001206827
tropy sampling 0.001191239
sampling heuristic 0.001191239
noise studies 0.001190521
class label 0.001185646
different picture 0.001176382
set size 0.001175854
human annotators 0.0011758279999999999
supervised classifier 0.001169565
results figure 0.0011494769999999999
test set 0.001144109
tion noise 0.001134782
entropy classifier 0.0011328290000000001
training 0.00113051
noise setting 0.001114806
supervised classification 0.0011123560000000001
random errors 0.0011082099999999999
incorrect labels 0.00110761
certain classifier 0.001101233
small set 0.0010929120000000001
same system 0.001087194
annotated instances 0.001083674
human annotator 0.001081358
systematic noise 0.001069268
large number 0.0010674740000000001
atic noise 0.001063225
classifier predictions 0.001061122
many nlp 0.001048873
human coder 0.00103374
machine translation 0.001020507
sense disambiguation 0.0010175409999999998
human coders 0.001010312
high accuracy 9.93821E-4
sense dis 9.762179999999999E-4
random choice 9.75118E-4
