Skip to main content

Table 3 Generalization performance

From: Prediction of breast cancer by profiling of urinary RNA metabolites using Support Vector Machine-based feature selection

Validation Sensitivity [%] Specifity [%] MCC* TP FN TN FP p-value
CV (10 fold) 83.5 90.6 0.743 71 14 77 8 2.2 × 10-16
LOO 83.5 85.9 0.694 71 14 73 12 2.2 × 10-16
  1. Prediction results for the best obtained feature selection. The p-value was computed by the Fishers' Exact Test.
  2. * for abbreviations refer to main text.