Label

A label is a target value or category attached to an example for training or evaluating a supervised model.

Plan for this page

Understand one concept in plain language and connect it to nearby ideas.

You will leave with: A label may be a class such as “spam,” a number such as a house price, or a structured annotation such as an object boundary. Labels encode human definitions, measurement processes, and sometimes disagreement or uncertainty.
Time: 3–8 min
Before you begin: No prerequisite. Keep one real example in mind.
Do this now: Read the explanation, then compare the example with the common misunderstanding.

Prepared next stepSee how examples shape behavior

LabelA label is a target value or category attached to an example for training or evaluating a supervised model.

Dataset
Feature
Prediction
Classification

Plain explanation

A label may be a class such as “spam,” a number such as a house price, or a structured annotation such as an object boundary. Labels encode human definitions, measurement processes, and sometimes disagreement or uncertainty.

Example

In a set of customer messages, reviewers might assign labels such as billing, technical problem, or account access.

Why it matters

A model can only learn the distinctions expressed by its labels and objective. Inconsistent, proxy, or contested labels can shape errors and unfair outcomes.

Common misunderstanding

Misconception: labels are neutral facts waiting inside the data. Correction: many labels are measured, interpreted, or negotiated and may contain error or judgment.

Related concepts

Follow relationships, not only the alphabet

DatasetA dataset is an organized collection of examples or records used for training, evaluation, analysis, or system operation.FeatureA feature is an input property or representation used by a model when producing an output.PredictionA prediction is a model’s estimated output for a given input, including a class, value, probability, sequence, or next element.ClassificationClassification is assigning an input to one or more defined categories, often with a score or probability for each.

See how examples shape behavior

Sources and scope

What supports this object

Machine Learning GlossaryGoogle for Developers

Defines labels in supervised learning; labeling practices and ambiguities depend on the domain and task.

Open source

Loading page content...

…