same entity 0.001819721
tweets information 0.001729345
semantic features 0.001693486
features ner 0.001658761
lexical features 0.001629035
syntactic features 0.001623906
related features 0.001547344
data set 0.001516278
entity types 0.0015136910000000002
entity type 0.001507722
orthographic features 0.001495289
entity dictionary 0.001492306
entity mention 0.0014885060000000001
different entity 0.001487138
thographic features 0.0014738849999999999
entity normalization 0.0014572270000000002
global information 0.001453286
multiple entity 0.001423407
test data 0.001414876
same tweet 0.001410125
same pos 0.001408332
nen model 0.001404536
entity recognition 0.001388602
entity variations 0.001362118
entity retrieval 0.001361751
rich entity 0.001358064
model con 0.0013397790000000001
entity track 0.001319346
entity recogni 0.001319346
annotated data 0.001311997
cal model 0.001311866
words vector 0.001293954
normalization information 0.0012929109999999999
crf model 0.0012840170000000001
news articles 0.001276355
same time 0.00127378
similar tweets 0.001258147
word pair 0.001257665
graphical model 0.001254586
news comments 0.0012540160000000002
features 0.00125246
data likelihood 0.001251705
word pairs 0.00124078
ner system 0.001231542
same set 0.001229609
data preparation 0.0012292050000000001
corresponding word 0.001211296
fresh information 0.0012009920000000001
nen system 0.001200977
cross tweets 0.001189255
system accuracy 0.001184335
stop word 0.001162639
frequency information 0.001159752
word prefixes 0.001148803
first tweet 0.001145452
english tweets 0.001123615
multiple tweets 0.001119768
related words 0.001107493
annotated tweets 0.001099268
entity 0.00109865
testing tweets 0.001086634
search results 0.00107781
unlabeled tweets 0.001075817
system per 0.001071369
system pre 0.001069864
words tim 0.001069216
system recog 0.0010568209999999999
stop words 0.001053431
labeled tweets 0.001053412
scale system 0.00105239
cascaded system 0.001050715
tweet normalization 0.001047631
short text 0.001043922
pos tagging 0.001043906
input tweets 0.001031909
model 0.0010288
cosine similarity 0.001027787
normalization method 0.001026748
same way 0.001025106
same product 0.001013877
feature set 0.001003728
single tweet 9.89781E-4
same group 9.76653E-4
second tweet 9.611140000000001E-4
same lemma 9.52229E-4
parameter set 9.49703E-4
pos tagger 9.48412E-4
search space 9.48377E-4
web text 9.473909999999999E-4
information 9.34334E-4
large number 9.3107E-4
advanced tweet 9.289880000000001E-4
learning approach 9.257E-4
weight set 9.20672E-4
feature weight 9.073239999999999E-4
random variables 9.00685E-4
method recog 8.99751E-4
web search 8.921739999999999E-4
method outper 8.91302E-4
random variable 8.90508E-4
