Skip to main content
. 2012 Mar 13;40(11):e87. doi: 10.1093/nar/gks230

Table 4.

Confusion matrix for the combined computational tool applied to a new set of sequences

Actual Class I Actual Class III Actual Class II
12 9 9
Predicted Class I 23 12 8 3
Predicted Class II 7 0 1 6

Class I: well-performing templates; Class II: poorly performing templates; Class III: templates with intermediate performance. Values represent the number of template sequences in each set.