liveBook · Manning

Gini impurity

As explained in Timeless Algorithms, The Foundational Ideas

Gini impurity measures the probability of misclassifying an observation when its class is assigned randomly according to the class distribution in a node.

Gini impurity, also called the Gini index, measures impurity rather than uncertainty in bits. It can be interpreted as the probability of misclassification when a node’s class is assigned randomly according to its class distribution. Decision trees use it to compare candidate splits and favor partitions with increasingly homogeneous outcomes. It often selects splits similar to entropy and can be computationally more efficient. section 5.3.1

How Timeless Algorithms, The Foundational Ideas covers it

How deep it goes
Contrasts impurity with entropy, noting similar split preferences and computational simplicity.
Examples
Uses loan-approval outcomes to illustrate nodes with balanced or dominant classes in the discussion of homogeneous splits.
What you can do after
Evaluate candidate decision-tree splits alongside entropy, expecting similar trees and little difference in predictive performance.
How it connects
Connects decision trees, entropy, and information gain through the goal of producing homogeneous groups.

Also explained in

Compare how 9 books explain Gini impurity →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Timeless Algorithms, The Foundational Ideas. The section links point to the parts of the book it draws on.