Skip to main content

Table 3 Accuracy of the random forest prediction rule in training and validation data

From: Gram-negative and -positive bacteria differentiation in blood culture samples by headspace volatile compound analysis

  Training set (n = 86) Validation set (n = 42)
Error rate (%) 9.1 16.7
Sensitivity (%) 97.5 93.3
Specificity (%) 74.8 58.3
AUC 0.93 0.89
  1. AUC area under the curve