Skip to main content

Table 6 Class specific F1-scores and overall weighted F1-score across all classes (in bold) from best performing model using different combinations of available feature sets

From: Machine learned daily life history classification using low frequency tracking data and automated modelling pipelines: application to North American waterfowl

 

All feature sets

Movement and timing and habitat features

Movement and timing and history features

Movement and timing features only

Brooding

0.552

0.240

0.000

0.000

Dead

1.000

0.992

0.997

0.984

Local

0.969

0.954

0.965

0.946

Migration

0.974

0.947

0.974

0.923

Molt-like

0.932

0.899

0.909

0.856

Molting

0.896

0.824

0.764

0.577

Nesting

0.919

0.899

0.897

0.792

Regional relocation

0.949

0.923

0.935

0.895

Weighted-F1

0.950

0.924

0.928

0.887