Skip to main content

Table 10 Stratified logistic regression feature ranking

From: Machine learning can predict survival of patients with heart failure from serum creatinine and ejection fraction alone

Rank

Clinical feature

Importance

1

Ejection fraction

4.13938106

2

Serum creatinine

3.69917184

3

Age

2.61938095

4

Creatinine phosphokinase

1.88929235

5

Sex

1.32038950

6

Platelets

1.06270364

7

High blood pressure

0.79478093

8

Anaemia

0.77547306

9

Smoking

0.65828165

10

Diabetes

0.60355319

11

Serum sodium

0.54241360

  1. Results of the feature ranking obtained by the stratified logisitc regression. Importance: coefficient of the trained logistic regression model, average of 100 execution
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us