A decision tree makes a prediction by asking a series of yes-or-no style questions about the data, like a flowchart, until it reaches an answer.
In plain words
The tree starts at the top with one question about a feature, such as whether a number is above some value. Each answer leads to another question, and each end point holds a prediction. The questions are learned from the training data, and the tree can predict categories or numbers.
A quick example
To guess whether a loan will be repaid, a tree might first ask whether income is above a threshold, then whether the person has missed payments before. Following the answers down the branches leads to an end point that says likely to repay or not.
Where you'll meet it
- Simple, explainable models for tabular data.
- As the building block of random forests and boosted trees.
Related
Sources
- DocsDecision trees, Google Decision Forests course · read 29 Sept 2026
- DocsDecision Trees, scikit-learn · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 29 Sept 2026. Spotted an error? Report it.