Decision Tree

lastmod 2026-07-31created 2024-04-245 edits

A Korean: 결정 트리 (decision tree) is a kind of decision support tool that diagrams decision rules and their outcomes as a tree structure. Decision trees are mainly used in operations research, especially in decision analysis, to find the strategy most likely to reach a goal.

A decision tree (decision tree) or tree model (tree model) is an analysis method that represents decision rules as a tree structure to classify (classification) the entire data set into several subgroups or to perform prediction (prediction).

Simple example

• Machine learning - 4. Decision Tree

A decision tree (Decision Tree, also called a decision-making tree) is a supervised learning model capable of both classification (Classification) and regression (Regression). A decision tree learns by asking a chain of yes/no questions, like a game of Twenty Questions. Suppose we want to distinguish a hawk, a penguin, a dolphin, and a bear. The hawk and the penguin have wings, while the dolphin and the bear do not. The question 'Does it have wings?' splits them into hawk, penguin / dolphin, bear. The hawk and the penguin can be split by the question 'Can it fly?', and the dolphin and the bear by the question 'Does it have fins?'. Below is a diagram of the decision tree.

Overview

A decision tree consists of three types of nodes:

• Decision node (decision node): usually shown as a square
• Chance node (chance node): usually shown as a circle
• End node (end node): usually shown as a triangle

See also

• Machine learning
• Regression (regression)
• Classification (classification)

Favorite site

• Decision tree - Wikipedia
• Decision tree learning - Wikipedia (ko), the free encyclopedia
• (ML) What is a Decision Tree?
• Decision Tree · ratsgo's blog
• Machine learning - 4. Decision Tree
• What is a Decision Tree? | IBM