Skip to main content

Table 3 Comparison of the random forest model with/without feature selection for the test set

From: Improving dengue fever predictions in Taiwan based on feature selection and random forests

 

No. of variables

Acc

Sen

Spec

AUC

Random forests (all variables included)

121

0.8742

0.9688

0.7302

0.9545

Random forests (top 60 variables selected)

60

0.8994

0.9479

0.8254

0.9547

  1. Abbreviations: Acc. Accuracy, Sen. Sensitivity, Spec. specificity, AUC area under the receiver operating characteristics curve
Back to article page

BMC Infectious Diseases

ISSN: 1471-2334

Contact us