Skip to main content

Advertisement

Table 6 The numbers of misclassifications for five binary classifiers using the support vector machine (SVM), random forest (RF) and logistic regression (LR) classification algorithms

From: Assessment of performance of survival prediction models for cancer prognosis

  Survival Threshold Risk Number of Training Number of Test samples Number of Misclassification
A B C D E
  Year Group samples
SVM 4 high 28 10 10 6 7 7 7
   low 50 9 1 2 3 2 3
  5 high 34 12 9 1 3 2 5
   low 44 7 1 2 2 2 2
  6 high 43 14 5 2 3 2 2
   low 35 5 1 1 2 1 1
RF 4 high 28 10 10 6 5 5 7
   low 50 9 2 1 3 1 3
  5 high 34 12 10 1 5 5 6
   low 44 7 1 2 2 0 2
  6 high 43 14 6 1 5 3 6
  low 35 5 2 1 2 1 2
LR 4 high 28 10 7 5 6 6 6
   low 50 9 1 2 3 4 4
  5 high 34 12 8 3 6 7 8
   low 44 7 0 4 2 3 1
  6 high 43 14 6 1 3 3 7
   low 35 5 0 3 2 2 1
  1. The binary classifiers are developed based on the 4-year, 5-year, and 6-year metastasis-free times to define the high and low risk classes.