Trees and ensembles · MLD-1156
Split criterion
A measure used to choose how a decision-tree node divides data.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Choose a threshold reducing impurity.
Original caveat
Greedy local choices do not guarantee a globally optimal tree.
Context and aliases
Level: Intermediate. Stages: Training · Inference · Evaluation.
Tags: ML.
Aliases: None recorded.
Related concepts
Original references
- scikit-learn: ensemble methods ↗ · Reference link
- scikit-learn: decision trees ↗ · Reference link
- Original reference URL ↗