Skip to main content

Table 6 Comparison of classifiers (Gaucher dataset)

From: Classification-based comparison of pre-processing methods for interpretation of mass spectrometry generated clinical datasets

  Ciphergen C Cromwell B
DLDA 77 (50–100) 73 (50–92)
Naive Bayes 73 (50–92) 68 (42–92)
PCDA 81 (58–100) 81 (50–100)
SVM 90 (75–100) 86 (67–100)
Tree 75 (42–92) 72 (42–92)
  1. Average classification accuracy (95% confidence interval) of Gaucher versus control on 500 test sets (size of training sets: 27, size of test sets: 12). Pre-processing was done using setting C for Ciphergen and setting B for Cromwell. Classifiers without feature selection.