juniorDimensionality Reduction
Why is feature scaling important for dimensionality reduction?
Updated May 16, 2026
Short answer
Scaling ensures all features contribute equally to distance and variance calculations.
Deep explanation
Techniques like PCA rely on variance and covariance, which are sensitive to feature scales. Without scaling, features with larger magnitudes dominate results.
Real-world example
In financial data, income would dominate age unless scaling is applied.
Common mistakes
- Applying PCA without normalization.
Follow-up questions
- Which scaling method is most common for PCA?
- Does t-SNE require scaling?