AI Dictionary › Fondamenti AI
Albero decisionale
A decision tree is a machine learning model that makes decisions by asking a sequence of simple questions about the data, organized in a tree shape: starting from a root, it goes down through nodes representing tests on a feature (is age over 30? does income exceed a threshold?) and arrives at a leaf containing the final prediction. It is arguably the most intuitive machine learning model, since it mirrors how a human would reason through successive categories.
During training, the algorithm looks, at each node, for the question that splits the data in the way most useful for separating classes or reducing error, using measures such as Gini impurity or entropy for classification, and variance for regression. The process repeats recursively on the resulting subsets until a stopping criterion is reached, such as a maximum depth or a minimum number of examples per leaf. A tree that is too deep tends to memorize the training data instead of generalizing, which is why pruning techniques are applied to cut away the least useful branches.
Decision trees are everywhere in applied AI precisely because of their transparency: credit scoring, medical triage, automated approval rules, simple recommendation systems. They are also the fundamental building block of more powerful models like random forest and gradient boosting, which combine many trees to achieve more accurate and stable predictions.
The idea of representing decisions as tree structures belongs to decision theory and predates computing; in machine learning the algorithm developed from the 1960s and 70s onward, with contributions such as CHAID and the systematization carried out by Leo Breiman and colleagues in the 1984 text Classification and Regression Trees, and by Ross Quinlan with the ID3 and C4.5 algorithms. The name simply describes the shape of the structure: a trunk that branches into increasingly specific decisions.
Grace's professional scenarios often present multi-level decision cases: understanding the logic of a decision tree helps you break a complex problem into simple, sequential questions when instructing an AI.
From our network
HSE Genius: AI for Safety Data Sheets
Extract SDS data, H phrases and ECHA compliance checks in seconds, powered by AI.
Visit hsegenius.com →From the Agora Intelligence blog
📱 Download the Android app (beta) iOS coming soon
Say what you mean. Get what you need.
Grace Certified, the AI coach that trains and certifies your prompt engineering, by Agora Intelligence.