Why standardize before PCA (usually)?
easyAnswer
- PCA maximizes variance in the original units.
- If features have wildly different scales (income in $ vs age in years), the high-variance one dominates → useless components.
- Standardize (z-score) so each contributes equally, unless: features are on the same natural scale (pixel intensities, log-returns) where scale carries meaning.
- Alternative: correlation-matrix PCA is equivalent to z-scored PCA.
Check yourself — multiple choice
- Never scale
- PCA maximizes variance in raw units → big-scale features dominate; standardize (or use correlation matrix) unless scale is meaningful
- Same as k-means
- Random
PCA scales-sensitive: standardize unless raw scale is meaningful.
#dimensionality-reduction#pca
Practise Unsupervised Learning
214 interview questions in this topic.