model features 0.003209053
syntactic features 0.003020039
baseline features 0.002920196
pcfg features 0.002783798
related features 0.002769905
stylistic features 0.002721377
stylometric features 0.002707984
tive features 0.002703463
motivated features 0.002686515
line features 0.002681704
sentiment features 0.0026492860000000003
features 0.00241901
feature score 0.002089001
feature analysis 0.002076857
language model 0.0020121030000000003
context language 0.001825353
language models 0.001766781
english language 0.001682341
feature 0.00154446
similar language 0.001543496
language styles 0.001534742
vandalism language 0.001508217
unique language 0.001508053
language processing 0.001493773
results data 0.001488311
contextual language 0.001479835
natural language 0.0014455470000000002
training corpus 0.001436683
test data 0.001297224
training set 0.001239904
slang words 0.0012311919999999999
language 0.00122206
vulgar words 0.001220441
ing data 0.00116518
guage model 0.0010880360000000001
different types 0.001085867
syntactic patterns 0.001076221
vandalism corpus 0.0010634989999999999
test set 0.001052244
previous work 0.001045907
pcfg parser 0.001043581
baseline classification 0.001038771
results table 0.001029231
annotated corpus 0.0010174189999999999
different aspects 0.001017172
baseline system 0.001007427
baseline approach 0.0010058620000000002
corpus potthast 0.0010020419999999999
class distribution 9.98872E-4
good score 9.939369999999999E-4
words 9.93877E-4
different stylometry 9.84863E-4
wikipedia edits 9.83234E-4
experimental results 9.77284E-4
related work 9.59554E-4
classification task 9.553840000000001E-4
wikipedia lan 9.5191E-4
wikipedia articles 9.48216E-4
detection system 9.45593E-4
training document 9.300580000000001E-4
first time 9.28417E-4
standard approach 9.146320000000001E-4
normal text 9.10737E-4
pcfg models 9.09509E-4
text difference 9.040999999999999E-4
context free 9.039E-4
parser cvandal 9.03589E-4
parser cregular 9.03589E-4
recent work 9.026209999999999E-4
future work 9.00958E-4
analysis table 8.9886E-4
new revision 8.97559E-4
edit distance 8.97091E-4
effective training 8.84607E-4
wikipedia vandalism 8.74976E-4
work wang 8.6732E-4
baseline classifier 8.671900000000001E-4
new pcfg 8.65722E-4
linguistic behavior 8.64344E-4
probabilistic context 8.62097E-4
first example 8.55646E-4
linguistic behav 8.54762E-4
classification rate 8.51583E-4
system description 8.50981E-4
lexical cues 8.50359E-4
wikipedia editors 8.48619E-4
wikipedia contin 8.473339999999999E-4
regular wikipedia 8.46994E-4
syntactic levels 8.467749999999999E-4
wikipedia authors 8.45692E-4
good edits 8.438110000000001E-4
guage models 8.427139999999999E-4
normal wikipedia 8.42265E-4
syntactic pat 8.37641E-4
previous revision 8.33873E-4
syntactic regularities 8.23327E-4
information gain 8.21998E-4
stylometric analysis 8.21371E-4
wikipedia users 8.20593E-4
sentiment wikipedia 8.190949999999999E-4
