Skip to main content

Table 3 Results of the secondary multivariate analysis focused on the ventilation modalities and the main outcome. In the table, listed are all the machine learning-tested models and their performances from the most performing to the least

From: Machine learning and predictive models: 2 years of Sars-CoV-2 pandemic in a single-center retrospective analysis

Model

AUC

CA

F1

Precision

Recall

Logistic regression

0.914

0.857

0.856

0.859

0.857

Neural network

0.913

0.846

0.845

0.845

0.846

Naïve Bayes

0.909

0.857

0.855

0.860

0.857

Random forest

0.907

0.847

0.846

0.849

0.847

Gradient boosting

0.899

0.844

0.843

0.844

0.844

SVM

0.880

0.849

0.843

0.844

0.844

CN2 rule inducer

0.843

0.774

0.774

0.775

0.774

SGD

0.836

0.849

0.847

0.853

0.849

Tree

0.774

0.786

0.787

0.789

0.786

kNN

0.593

0.589

0.583

0.582

0.589