Skip to main content

Table 5 Average BRCA subtype classification performance results under different confidence thresholds for pseudo-labeling based on Tenfold cross-validation

From: meth-SemiCancer: a cancer subtype classification framework via semi-supervised learning utilizing DNA methylation profiles

Confidence threshold

0.5

0.6

0.7

0.8

0.9

Accuracy

0.836

0.839

0.842

0.844

0.842

F1-score

0.821

0.826

0.828

0.834

0.831

MCC

0.747

0.743

0.753

0.751

0.745

Recall

0.729

0.719

0.719

0.729

0.694

Precision

0.793

0.782

0.794

0.798

0.768

Kappa

0.742

0.737

0.746

0.742

0.735