Beginner · Data

Annotation

Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.

TL;DR. The act of attaching labels, tags, or structured information to raw data.

Technical Definition

The act of attaching labels, tags, or structured information to raw data.

How it works

Annotation can be simple (a class per image) or rich (segmentation masks, dialogue intent, RLHF preferences). Good annotation requires clear guidelines, multiple annotators per item for agreement, and rigorous quality control. Bad annotation is the silent killer of ML projects.

Related Concepts

• Training Data — The portion of a dataset used to fit a model's parameters.
• Label — The correct answer attached to a training example in supervised learning.
• Human-in-the-loop — System designs where humans review, correct, or guide AI outputs as part of the workflow.