Skip to main content
. 2011 Dec 22;6(12):e29373. doi: 10.1371/journal.pone.0029373

Table 4. Number of species from the 168 species classified in different accuracy classes of AUC and TSS based on two modelling techniques.

Predictor variables Climate Vegetation Landscape
Evaluation method AUC TSS AUC TSS AUC TSS
Model technique BRT RF BRT RF BRT RF BRT RF BRT RF BRT RF
High-performance 28 35 17 19 23 27 15 21 17 28 5 10
Good-performance 61 78 28 32 53 60 21 25 35 71 19 30
Fair-performance 70 53 79 90 78 46 80 74 75 59 66 85
Poor-performance 9 2 44 27 14 28 51 48 41 5 78 43
Fail 0 0 0 0 0 7 1 0 0 5 0 0

AUC: High = AUC>0.9, Good = 0.9<AUC<0.8; Fair = 0.7<AUC<0.8; Poor = 0.6<AUC<0.7. Fail AUC<0.6.

TSS: High = TSS>0.8, Good = 0.8<TSS<0.6; Fair = 0.6<TSS<0.4 and Poor = 0.2<TSS<0.4. Fail TSS<0.4.