Random Forest

Feature Importance

Metric calculated in Random Forest that evaluates the contribution of each variable to model accuracy, typically measured by the mean decrease in impurity (Gini or entropy) or the increase in OOB error when the variable is permuted.

← Tillbaka