training data 0.001884559
previous model 0.001851286
model parameters 0.001829926
regression model 0.001821191
data set 0.001786512
bilinear model 0.001781337
predictive model 0.001731892
different data 0.0017248889999999998
ear model 0.001699637
curate model 0.001694142
training set 0.001579991
user information 0.001545555
first data 0.001514759
model 0.0014953
word features 0.0014841720000000002
new data 0.001439965
input data 0.001436795
learning task 0.001436382
feature selection 0.001425111
different set 0.001420321
different task 0.001413725
feature space 0.001382655
test set 0.00135445
second data 0.001345544
data processing 0.00129842
data sets 0.001275839
cau data 0.001244313
different users 0.0012291889999999999
linear learning 0.001185464
topic word 0.001178482
political word 0.001171448
same set 0.001170111
different words 0.0011543389999999999
training process 0.001151579
feature 0.00114415
many features 0.001138544
large performance 0.0011327109999999998
learning problem 0.001112615
user weight 0.001111812
word match 0.001110897
training inputs 0.001094987
user score 0.0010916049999999998
performance results 0.001081225
user weights 0.001081084
simple features 0.001078146
training complexity 0.001076348
training sets 0.001069318
training instances 0.001053478
linear models 0.0010514259999999998
word lists 0.001050869
training points 0.00105037
validation set 0.001047832
modest training 0.00104265
user modelling 0.001039738
training loop 0.001037661
erative training 0.001037661
task domain 0.001022726
sparse set 0.001013812
user space 0.001010033
different term 0.001009775
user table 0.001006445
loss function 0.001003128
standard linear 0.001000438
fixed set 0.001000391
strong performance 9.99057E-4
learning functions 9.98022E-4
task domains 9.90208E-4
user fpo 9.892219999999999E-4
first time 9.87296E-4
english language 9.86691E-4
valuable information 9.778760000000001E-4
optimisation task 9.77404E-4
user opinions 9.7721E-4
word frequencies 9.765960000000001E-4
user spaces 9.7606E-4
user groups 9.73955E-4
unstructured user 9.718509999999999E-4
many users 9.71491E-4
ture set 9.70589E-4
shared set 9.68084E-4
word frequen 9.66962E-4
test point 9.647169999999999E-4
consistent set 9.58844E-4
average performance 9.58836E-4
same time 9.472160000000001E-4
different types 9.43505E-4
performance improvement 9.397889999999999E-4
ing users 9.29356E-4
informative features 9.26864E-4
learning property 9.245200000000001E-4
previous models 9.239539999999999E-4
other loss 9.23679E-4
biconvex learning 9.05496E-4
input matrix 9.041330000000001E-4
learning schemes 9.01336E-4
arate learning 9.01336E-4
different characteristics 8.94241E-4
regression models 8.938589999999999E-4
different magnitude 8.85879E-4
inference performance 8.83219E-4
