Skip to main content

Table 5 Comparison of AUCs. The upper part of the matrix represents the average AUC differences between models. The lower part represents statistical significance (p values) of paired t-tests

From: Comparison and development of machine learning tools in the prediction of chronic kidney disease progression

pVal\fold change

LR

Elastic Net

Lasso

Ridge

SVM

RF

k-NN

NN

XGBoost

LR

–

0.002

0.001

0.008

0.017

0.019

0.071

0.019

0.005

Elastic Net

0.195

–

− 0.001

0.006

0.015

0.017

0.069

0.017

0.003

Lasso

0.489

0.372

–

0.007

0.016

0.018

0.070

0.019

0.004

Ridge

0.001

0.014

0.001

–

0.009

0.011

0.063

0.011

− 0.003

SVM

0.000

0.001

0.000

0.055

–

0.002

0.054

0.003

− 0.012

RF

0.000

0.001

0.000

0.019

0.536

–

0.052

0.000

− 0.014

k-NN

0.000

0.000

0.000

0.000

0.000

0.000

–

− 0.052

0.066

NN

0.000

0.004

0.002

0.046

0.540

0.956

0.000

–

− 0.015

XGBoost

0.219

0.488

0.316

0.488

0.010

0.000

0.000

0.015

–