Skip to main content

Table 3 Effect of variable selection on model performance

From: Predicting moose behaviors from tri-axial accelerometer data using a supervised classification algorithm

 

Full RF

Reduced RF

 

Recall

Precision

Recall

Precision

Foraging

0.86

0.90

0.84

0.89

Lying_o

0.78

0.34

0.75

0.41

Lying_u

0.74

0.84

0.74

0.83

Ruminate

0.79

0.80

0.82

0.78

Run

0.74

0.28

0.73

0.24

Stand

0.55

0.56

0.52

0.53

Walk

0.79

0.62

0.79

0.60

  1. Comparison of model performance between the full random forest model run with all 50 predictor variables and subsequent reduced random forest model run with only the 16 most important variables
Back to article page

Animal Biotelemetry

ISSN: 2050-3385