Random Forest

A random forest uses many decision trees. Each tree votes, and the answer with the most votes wins.

What it does

A random forest uses many decision trees. Each tree votes, and the answer with the most votes wins.

Ensemble: A team of models • Voting: Picking the common result

How it works

• Build many small decision trees.
• Let each tree look at a slightly different part of the data.
• Count the votes and choose the most popular answer.

Uses and applications

• A movie app can combine many small guesses to recommend a movie.
• A bank can use many checks together to help spot risky transactions.
• A doctor’s tool can combine many clues to help sort scans for review.

In short

It is like asking many friends for advice. One friend may be wrong, but a big group vote is often better.

Related methods

• Decision Tree: The single model that a random forest repeats many times.
• Gradient Boosting: Also combines trees, but builds them in sequence rather than independently.
• Isolation Forest: Uses random trees to spot unusual points instead of predicting a label.

Back to Prediction and classification · All topics

The interactive demonstration and the Python walkthrough for this topic need JavaScript.