Table 3 Matrix of accuracy results, according to different scales and algorithms.

From: Machine learning model from a Spanish cohort for prediction of SARS-COV-2 mortality risk and critical patients

Scaler

SVM

LR

K-neighbors

Decision Tree

Naive Bayes

Random Forest

MLP

GP

AdaBoost

Bagging

MinMax

0.8826

0.8876

0.8759

0.8776

0.5378

0.8901

0.8901

0.8718

0.886

0.8793

Standard

0.8868

0.8968

0.8818

0.8801

0.5378

0.8976

0.8926

0.8718

0.8843

0.8859

MaxAbs

0.8826

0.8876

0.8759

0.8751

0.5378

0.8943

0.8909

0.8718

0.8851

0.8793

Robust

0.8826

0.8968

0.8784

0.8743

0.5378

0.8868

0.8934

0.8726

0.8835

0.8818

Quant-Normal

0.8859

0.8968

0.8843

0.8826

0.5378

0.8951

0.8993

0.8734

0.8843

0.8918

Quant-Uniform

0.8809

0.8876

0.8759

0.8693

0.5378

0.8984

0.8893

0.8718

0.886

0.8793

PowerTransf-yeoJhonson

0.8818

0.8968

0.8776

0.8693

0.5378

0.8935

0.8951

0.8718

0.8843

0.8859

Back to article page