What are the potential drawbacks of using PCA for dimensionality reduction?

  • It always improves model performance
  • It can lead to information loss and doesn't consider class labels
  • It normalizes the variance of the data
  • It removes all noise and outliers
The potential drawbacks of using PCA include the risk of information loss since it only considers variance, not class labels, and might remove meaningful information that doesn't align with the directions of maximum variance.
Add your answer
Loading...

Leave a comment

Your email address will not be published. Required fields are marked *