training data 0.00295983
test data 0.0028446969999999998
unlabeled data 0.00271193
other data 0.002687857
corpus data 0.002657372
data corpus 0.002657372
data set 0.002588048
ing data 0.002544823
data sets 0.002439372
target data 0.0024384569999999998
rect data 0.002308155
similarity graph 0.0016892699999999999
category graph 0.001659435
test document 0.001504339
learning method 0.0014927410000000001
graph con 0.001464607
undirected graph 0.00142887
graph construction 0.001415836
proper graph 0.001413727
good graph 0.001388799
adjacency graph 0.0013760559999999999
complete graph 0.001356964
ity graph 0.0013563759999999998
graph propaga 0.0013532549999999998
ilarity graph 0.001351866
training documents 0.001319016
document label 0.001288679
tion algorithm 0.001236868
test documents 0.001203883
categorization algorithm 0.001192196
pagerank algorithm 0.0011811529999999999
propagation method 0.001172402
ank algorithm 0.001163491
egorization algorithm 0.001163491
quality training 0.001147303
document vector 0.001144088
training datum 0.001142666
many document 0.001135193
value figure 0.001127653
latent information 0.0011240640000000001
category label 0.0011206620000000001
neighbors method 0.001106629
category category 0.00110641
graph 0.00110623
common test 0.0011055
surface information 0.00108134
summarization method 0.001079001
own method 0.001054716
topicrank method 0.00104883
method topicrank 0.00104883
document summarization 0.001033843
important sentences 0.00102857
gbssl method 0.001015831
gation method 0.001012848
face information 0.0010128350000000001
document vectors 9.9719E-4
document categorization 9.9709E-4
experimental results 9.78119E-4
ping learning 9.75514E-4
label propagation 9.734790000000001E-4
document catego 9.67456E-4
similarity measure 9.407889999999999E-4
other cases 9.31575E-4
natural language 9.27125E-4
topic categories 9.236839999999999E-4
algorithm 9.16328E-4
other hand 9.1028E-4
crude figure 9.00515E-4
training 8.9825E-4
estimation value 8.97784E-4
english news 8.89176E-4
objective function 8.88957E-4
same cate 8.88639E-4
interest figure 8.87895E-4
grain figure 8.808189999999999E-4
trade figure 8.808189999999999E-4
ship figure 8.808189999999999E-4
wheat figure 8.808189999999999E-4
corn figure 8.808189999999999E-4
acq figure 8.808189999999999E-4
whole category 8.51453E-4
first term 8.44533E-4
text body 8.33112E-4
relative value 8.3139E-4
correct value 8.30398E-4
sine similarity 8.2913E-4
first con 8.15987E-4
label propa 8.126450000000001E-4
respective categories 8.052269999999999E-4
target documents 7.97643E-4
ral language 7.93047E-4
main clustering 7.897939999999999E-4
large quantity 7.871919999999999E-4
rization problems 7.860529999999999E-4
language processings 7.826459999999999E-4
language lexicons 7.826459999999999E-4
maximum value 7.80632E-4
topic estimation 7.73245E-4
average values 7.72694E-4
information 7.67153E-4
