Label
The known answer for each training example, what a supervised model is trying to predict.
- Can be discrete (class) or continuous (number).
- Label quality is the ceiling on model quality.
- Noisy or biased labels propagate directly into predictions.
When to use: Required for every supervised problem; without labels, only unsupervised methods apply.
Example: Sentiment classifier labels are {positive, neutral, negative} per review. For a house price model, the label is the actual sale price.