Skip to main content

Table 2 Holm–Bonferroni post hoc test across machine learning models performed using age in the original format and logarithmic autoantibody levels

From: Development and validation of machine learning-based risk prediction models of oral squamous cell carcinoma using salivary autoantibody biomarkers

Algorithm (AUC)

LR 0.784 ± 0.057

RF 0.764 ± 0.057

SVM 0.772 ± 0.059

XGBoost 0.778 ± 0.051

Stacking 0.795 ± 0.055

RF

< 0.001

–

–

–

–

SVM

0.006

0.188

–

–

–

XGBoost

0.270

0.002

0.270

–

–

Stacking

0.001

 < 0.001

 < 0.001

 < 0.001

–

  1. LR logistic regression, RF random forest, SVM support vector machine with a radial basis function kernel, XGBoost eXtreme Gradient Boosting, Stacking a stacking model that contained all models, SD standard deviation
Back to article page

BMC Oral Health

ISSN: 1472-6831

Contact us