Possible Duplicate:
What is the intuitive relationship between SVD and PCA

I am confused between PCA and SVD.

The wikipedia page for the PCA has this line:

"PCA can be done by eigenvalue decomposition of a data covariance matrix or singular value decomposition of a data matrix, usually after mean centering the data for each attribute."

Does this mean that PCA = SVD of a data matrix?

Is there an article/tutorial that explains the difference?


marked as duplicate by J. M. is a poor mathematician, Srivatsan, Rudy the Reindeer, robjohn, t.b. Dec 13 '11 at 14:27

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.


PCA needs SVD or Eigenvalue decomposition to calculate principal components (PCs). That is, PCs are the result of SVD of the samples.

PCA has two step.

  1. derive PCs which are axes of PCA using SVD and select n eigenvectors corresponding to the n largest eigenvalues.
  2. project some samples to these PCs (to denoise, dimensionality reduction, feature extraction)

You can find some articles using the keyword "tutorial principal component analysis".

Hope this helps


Not the answer you're looking for? Browse other questions tagged or ask your own question.