news data 0.001790524
new model 0.0015582410000000001
news corpus 0.001540409
baseline model 0.001531783
full model 0.001510343
unsupervised model 0.001476299
model wtmf 0.001436554
wtmf model 0.001436554
based model 0.001433706
variable model 0.001413381
novel model 0.001406883
model minus 0.0013708750000000001
final model 0.001363854
time news 0.001358356
word level 0.001358066
retrieval model 0.001344865
word vector 0.0013439419999999999
other features 0.001343809
robust model 0.001340307
news task 0.001328474
standard data 0.001301887
news nodes 0.001281536
word distribution 0.001281342
training corpus 0.001270022
context information 0.0012532139999999999
news document 0.001246639
topic information 0.001246339
correct news 0.001204712
news recommendation 0.001199171
occurrence information 0.001197703
temporal information 0.001197065
word pair 0.001195476
word links 0.001194209
additional information 0.001194033
input data 0.001192206
news domain 0.0011884859999999999
news subgraph 0.001183337
death news 0.001178319
same time 0.0011764710000000001
surface word 0.001172528
news articles 0.001170847
shared information 0.001170338
nlp features 0.001167609
news article 0.00116139
candidate news 0.001160277
news entry 0.001155043
breaking news 0.001152337
related news 0.001152165
natural language 0.001131764
word matching 0.001130938
relevant news 0.0011283019999999999
face word 0.001125121
text vector 0.001125068
rect news 0.00112237
news arti 0.001121257
data points 0.001119995
news genre 0.00111949
model 0.00111881
authorship information 0.001118392
news summary 0.001118252
news title 0.00111675
news titles 0.00111675
news summaries 0.00111675
irrelevant news 0.00111675
nytimes news 0.00111675
news recommen 0.00111675
same event 0.001115073
data sparseness 0.001113451
sentiment information 0.001108335
link information 0.0011073909999999999
information characteristi 0.0011004539999999998
valuable information 0.0011004539999999998
poral information 0.0011004539999999998
same hashtag 0.0010916670000000002
same topic 0.001082234
same corpora 0.0010766320000000001
different number 0.0010762179999999999
text nodes 0.001070597
hashtag words 0.001068214
training set 0.001064377
short text 0.001061834
results table 0.0010552640000000002
language understanding 0.001054379
same way 0.0010394570000000001
text similarity 0.0010367549999999999
pos tagger 0.00102465
different values 0.001021901
url features 0.0010207430000000002
other tweets 0.001012809
additional text 0.0010008740000000001
words weight 9.975119999999999E-4
test set 9.96254E-4
specific features 9.90264E-4
features wtmf 9.84344E-4
text modeling 9.84036E-4
same entities 9.79185E-4
text links 9.75335E-4
other metrics 9.73142E-4
key words 9.464899999999999E-4
text semantics 9.41901E-4
