Skip to main content

Table 3 Confusion matrix of S. peruvianum and S. corneliomulleri in the AFLP dataset

From: Reliable genomic strategies for species classification of plant genetic resources

 

RF

NB

NJ

A priori classification

S. per

S. cor

Other

S. per

S. cor

Other

S. per

S. cor

Other

S. peruvianum (n = 12)

0.92

0.08

0.00

0.00

0.00

1.00

0.75

0.25

0.00

S. corneliomulleri (n = 4)

1.00

0.00

0.00

0.00

0.25

0.75

0.50

0.50

0.00

 

1-NN

3-NN

A priori classification

S. per

S. cor

Other

S. per

S. cor

Other

S. peruvianum (n = 12)

0.75

0.25

0.00

1.00

0.00

0.00

S. corneliomulleri (n = 4)

0.75

0.25

0.00

0.25

0.67

0.25

  1. Confusion matrix of the AFLP tomato dataset, showing fractionally how often S. peruvianum and S. corneliomulleri are classified as themselves, as each other, and as other species by Random Forest (RF), Naive Bayes (NB), Neighbour-Joining (NJ), 1-Nearest Neighbour (1-NN), and 3-Nearest Neighbours (3-NN)