Question : What is Principal component Analysis (PCA) ?
Answer:  Principal Component Analysis is a statistical process that reduces the dimension of the original data by  preserving the most important information while removing the less relevant information. It does so by projecting the original data into the orthogonal space and then considering only the dimensions that capture the most variance or most information.

Question : Can we visualise PCA in practice with a  simple example?
Answer: PCA does dimension reduction in following steps

  • Find the rotation of the data, such that the first dimension captures the most variation in the data and the second dimension captures the next most variation.
  •  Project the original data into this new dimension.
  • Reduce the dimension by  disregarding the dimensions that  do not capture a lot of information

PCA - dimesionality reduction.JPG