Decision Trees
Long-tail route · model intuition
Use this when split logic should feel earned instead of abstract.
MLU scrollytelling lesson that builds a decision tree step-by-step and links entropy/information gain math to split decisions.
At a glance
Live interactive
The maintained version lives at the route linked here, with this garden note keeping the embed and editorial framing.
Best companion
Pair it with [[Random-Forest]] once single-tree logic feels clear and you want to see how bagging changes the bias-variance tradeoff.
Time and level
About 10 to 15 minutes. Beginner friendly. Best if you want one model that is still readable node by node.
Reading path
- Open the live interactive: https://kohnnn.github.io/interactive-explanation/decision-trees/
- Continue through Random-Forest when you want the ensemble follow-up after the single-tree logic clicks
- Move through the local archive via Interactive or Visual Notes