Skip to main content

Table 3 Average error rates over 100 runs on features from four HCV genome regions using two different classification algorithms.

From: HCV genotyping using statistical classification approach

Classification Method

Region on HCV Genome

5' NCR

CORE

E1

NS5B

SVM

21.98

19.66

1.60

0.21

Random Forest

24.28

3.98

0.56

0.19

  1. Error rates are computed as average error rates over 100 runs, that is, a cross-validation procedure of training on 90% of the data and testing on the remaining 10% was repeated 100 times and the errors averaged.