What is PCA in unsupervised learning?

Updated May 15, 2026

Short answer

PCA reduces dimensionality by projecting data into principal components.

Deep explanation

It identifies directions of maximum variance using eigenvectors of covariance matrix.

Real-world example

Reducing image dimensions for faster processing.

Common mistakes

  • Assuming PCA preserves interpretability of features.

Follow-up questions

  • What are eigenvalues in PCA?
  • When should PCA be used?

More Unsupervised Learning interview questions

View all →