Data Science & ML
Intermediate

Gini Impurity

Probability of misclassifying a random element; used in decision trees.

Formula

Gini=1−∑i=1npi2Gini = 1 - \sum_{i=1}^{n} p_i^2

Variables

p_iClass proportion
nClasses

Example

Pure node (p=1): Gini = 0

Did You Know?

CART uses Gini impurity while ID3 uses entropy.

decision-treeimpurity