other term 0.002874164
term set 0.002656951
many term 0.002607314
term selection 0.002606214
large term 0.002602771
term weighting 0.0025947889999999997
small term 0.0025444319999999997
term subset 0.0025286889999999998
candidate term 0.0024938399999999998
term subsets 0.0024659499999999997
term sets 0.002461736
optimal term 0.002459242
whole term 0.002450523
term sub 0.002430336
term selec 0.002422017
term density 0.0024205389999999998
scalable term 0.002418063
aggressive term 0.002410239
term rankings 0.002410239
gressive term 0.002410239
bigram term 0.002410239
different document 0.001905104
document vector 0.001829851
document frequency 0.001703159
english document 0.001627734
document number 0.00162596
document collection 0.001527868
document index 0.0015231210000000001
document indexing 0.0015138830000000002
document sparseness 0.001499721
different terms 0.0014958060000000001
total document 0.001492284
space model 0.001470516
document collections 0.0014640550000000001
primary document 0.001434868
document sizes 0.001426531
document hold 0.001426531
frequency terms 0.0012938609999999999
distribution figure 0.0012519599999999999
document 0.0011941
different target 0.001191521
length figure 0.0011774110000000002
log probability 0.001140809
same ratio 0.001096495
model 0.00109647
vector length 0.0010924810000000002
same avl 0.0010835250000000001
sts figure 0.001073279
different dimensionalities 0.001033705
corpus avl 0.001031924
irrelevant terms 0.001021953
other experiment 0.001017374
different docu 0.0010114899999999999
vector space 0.001009797
other implementations 0.001007609
text information 0.001005545
high performance 0.001004071
probability ratio 9.98595E-4
average vector 9.9141E-4
different dimen 9.865149999999999E-4
chinese corpus 9.834359999999999E-4
other pattern 9.812179999999998E-4
few training 9.79296E-4
same value 9.77169E-4
other settings 9.77059E-4
other metrics 9.67487E-4
same experiments 9.64222E-4
other cases 9.629949999999999E-4
different doc 9.586589999999999E-4
low performance 9.52949E-4
different dimension 9.47714E-4
english text 9.47702E-4
sample vector 9.44999E-4
different grayness 9.440189999999999E-4
performance comparison 9.342420000000001E-4
many documents 9.2588E-4
first step 9.22579E-4
performance improvements 9.19546E-4
same ranking 9.19275E-4
high frequency 9.163069999999999E-4
chinese text 9.08163E-4
training samples 9.01745E-4
actual performance 8.98943E-4
text categorization 8.86671E-4
vector machines 8.866570000000001E-4
selection method 8.85484E-4
vector lengths 8.83079E-4
ing set 8.76189E-4
close performance 8.71038E-4
support vector 8.70704E-4
empirical avl 8.695879999999999E-4
logarithmic vector 8.679130000000001E-4
erage vector 8.679130000000001E-4
learning process 8.632970000000001E-4
feature selection 8.43233E-4
target dimensionality 8.40928E-4
test set 8.36254E-4
continuous function 8.3478E-4
monotone function 8.33542E-4
performance curve 8.31246E-4
