Encyclopedia · 176 concepts

Core ML Concepts · beginner · concept 13 of 176

Decision Trees & Random Forests

A decision tree splits data with a cascade of if-then questions; a random forest averages hundreds of trees trained on random slices of the data. On tabular business data they remain brutally hard to beat, and unlike neural networks a single tree can be read and explained line by line.

Key terms

Splitting criterionGini impurityEnsembleBaggingFeature importance

Learn these first

Where you meet it in the real world

Credit scoring, churn prediction, fraud flags, any tabular dataset in industry