Table 4 Evaluation metrics for machine learning Erigeron spp. and Solidago spp. presence-absence classification models (GBM, SVM, RF) fitted with fused field and remote sensing data; the highest values for each genus are in bold (not tested for differences).

From: Predicting the co-invasion of two Asteraceae plant genera in post-mining landscapes using satellite remote sensing and airborne LiDAR

 

Erigeron

Solidago

GBM

SVM

RF

GBM

SVM

RF

AUC (± SD)

0.923 (± 0.054)

0.921 (± 0.051)

0.920 (± 0.050)

0.768 (± 0.083)

0.795 (± 0.079)

0.782 (± 0.082)

Sensitivity

0.829

0.771

0.857

0.656

0.594

0.563

Specificity

0.925

0.906

0.943

0.857

0.857

0.839

Precision

0.879

0.844

0.909

0.724

0.704

0.667

F1

0.853

0.806

0.882

0.689

0.644

0.610

Accuracy

0.886

0.852

0.909

0.784

0.761

0.739

Back to article page