training data 0.003352129
test data 0.0031942430000000003
annotated data 0.002941663
data set 0.002938482
data sets 0.002871465
general data 0.002864648
data size 0.0028527910000000004
ing data 0.00285017
specific data 0.002847473
statistical data 0.002812523
data selection 0.002796551
statistic data 0.0026970830000000003
domain ner 0.002187742
domain information 0.002140334
different domain 0.002007872
feature vector 0.002003419
ner model 0.001982732
ner performance 0.001882805
dependency feature 0.001774351
chinese word 0.001699275
specific domain 0.001695683
word segmentation 0.0016904749999999999
new domain 0.0016641919999999999
domain tags 0.0016572269999999998
corresponding domain 0.001623048
ner system 0.001591014
economic domain 0.001588794
document domain 0.001587998
ferent domain 0.001584816
probability model 0.001584751
language model 0.001581
domain per 0.001580951
politics domain 0.001572149
domain dependency 0.001571341
learning approach 0.001568865
domain categories 0.00156321
domain informa 0.001547109
seven domain 0.0015454969999999998
nomic domain 0.0015436109999999999
tainment domain 0.0015436109999999999
recognition model 0.001536036
chinese ner 0.001517377
general model 0.001507848
feature 0.00149994
specific model 0.0014906729999999999
segmentation information 0.001461169
linguistic features 0.001436499
markov model 0.001427457
distribution features 0.001386844
chinese words 0.0013761770000000001
model per 0.0013759409999999999
ner task 0.0013733360000000002
ner models 0.001371629
word patterns 0.0013680889999999999
bility model 0.0013387219999999999
final performance 0.001334225
training sets 0.001326154
general training 0.0013193369999999999
initial performance 0.001311738
general ner 0.00130674
performance stability 0.001304809
english ner 0.0013041530000000002
domain 0.00129693
specific ner 0.0012895650000000001
performance curves 0.001258853
tic features 0.001256014
ner systems 0.00125303
performance curve 0.001245999
tribution features 0.001245305
performance enhancement 0.001242343
different domains 0.001239816
performance sta 0.001239772
performance degrading 0.001239772
corresponding training 0.001229527
classification method 0.001218132
ner mod 0.001199535
information extraction 0.001185542
ner perfor 0.001184306
applicable ner 0.001177549
much training 0.001171701
test sets 0.001168268
general test 0.001161451
learning methods 0.001154753
recognition system 0.001144318
specific test 0.001144276
trained ner 0.001142952
possible class 0.001140365
stable ner 0.001137014
annotated corpus 0.001118577
experimental results 0.001108036
supervised learning 0.001106973
different distribution 0.001100417
model 0.00109192
information integration 0.001091631
learning approaches 0.0010836539999999999
machine learning 0.001080377
selection method 0.001064681
minimization classifier 0.001032579
class labels 0.0010300700000000001
corpus size 0.0010297050000000001
