Skip to main content

Table 1 Results from a 10-fold cross validation. Classification errors (% misclassified) for the different methods at their optimal word length and for various data sets. Singleton genera errors are not included since they add the same to all methods

From: Comparing K-mer based methods for improved classification of 16S sequences

 

Trainingset9

SilvaSet

Method

Full-length

Fragments

Full-length

Fragments

PLSNN

4.15 (K7)

16.96 (K8)

4.87 (K8)

24.33 (K7)

Multinomial

4.70 (K8)

16.00 (K8)

5.68 (K8)

19.73 (K8)

NN

4.99 (K7)

16.54 (K8)

5.63 (K8)

24.02 (K8)

RDP

5.43 (K8)

16.42 (K8)

6.55 (K8)

20.49 (K8)

Markov

5.93 (K4)

21.78 (K6)

8.10 (K6)

22.98 (K7)