PCA and FACTOR ANALYSIS

[RohitDhankar]

PCA and FACTOR ANALYSIS , being the SuperSet within which one common method is PCA .. https://en.wikipedia.org/wiki/Factor_analysis

1/ What best options for Dimensionality Reduction - https://en.wikipedia.org/wiki/Dimensionality_reduction

2/ Feature Selection Options -
a) filter strategy (e.g. information gain)
b) wrapper strategy (e.g. search guided by accuracy)
c) embedded strategy (selected features add or are removed while building the model based on prediction errors).

3/ Linear data transformation---

a) PCA -- Principal Component Analysis, -- Performs linear mapping of data to a lower-dimensional space ( Reduces COUNT of Independent Variables ) in such a way that - variance of the data in the low-dimensional representation is maximized. The new set of Independent Variables which is achieved as a result of the PCA are less in number but explain the Variance of the data better.

b) Other options for Linear data transformation ..

4/ Non-Linear data transformation---