Inference, estimation, and decision-making from data
A dataset often mixes very different kinds of measurements: age in years, income in thousands, distance in millimetres. Their raw numbers can differ by enormous factors even when every feature carries equally useful information. Feature scaling puts these measurements on comparable numerical scales, so an algorithm judges them by what they mean, not by how big their units happen to be.
🔒 This is a Pro lesson — the interactive figure, worked examples, quiz and practice open with Pro access.
▶ Feature Scaling & Standardization