Table 1 Prediction accuracy in validation set of individuals

From: A highly accurate risk factor-based XGBoost multiethnic model for identifying patients with skin cancer

Dataset

Positive predictive value

Negative predictive value

Sensitivity

Specificity

Precision

Recall

F1

All

0.80882353

0.99967187

0.99337748

0.98849625

0.80882353

0.99337748

0.89165091

Male

0.80712166

0.99976744

0.996337

0.98510541

0.80712166

0.996337

0.89180328

Female

0.81049563

0.99961208

0.99049881

0.99062951

0.81049563

0.99049881

0.89150187

EUR

0.82481752

0.99978101

0.99728445

0.98320233

0.82481752

0.99728445

0.90288875

not EUR

0.6988417

0.99954119

0.96276596

0.99491127

0.6988417

0.96276596

0.8098434

  1. Prediction accuracy of XGBoost multiethnic model (any skin cancer, all participants and both sexes) in the validation set of individuals. Shown are PPV, NPV, sensitivity, specificity, precision, recall and F1 statistic. The results are shown when considering All participants, Males only, Females only, EUR only, and non-EUR only.
Back to article page