Skip to main content

Table 4 Predictive abilities of the machine learning models for the asthma diagnosis after cross-validation with Bayesian hyperparameter tuning

From: Increasing the accuracy of the asthma diagnosis using an operational definition for asthma and a machine learning method

Model

Accuracy

AUC

Sensitivity

(Recall)

Specificity

(Precision)

Tree

0.8353

0.8712

0.8070

0.9388

Random forest

0.7412

0.8835

0.6140

1.0000

XGBoost

0.8706

0.9301

0.8246

0.9792

LGBM

0.8118

0.8966

0.7193

1.0000

CatBoost

0.8000

0.9029

0.7544

0.9348

  1. LGBM Light gradient boosting model, XGBoost Extreme gradient boosting
Back to article page

BMC Pulmonary Medicine

ISSN: 1471-2466

Contact us