Which are linear combinations the original features

Solved Step by Step With Explanation- PCA vs MLDA

Questions

Principal Component Analysis (PCA) and Multiple Linear Discriminant Analysis (MLDA) are both dimensionality reduction techniques used in the field of machine learning and statistics. However, they serve different purposes and have distinct approaches:

1. Purpose:
2. • MLDA: MLDA is supervised because it considers class labels. It aims to find a linear transformation that best separates different classes while minimizing the variance within each class.

3. Objective:
4. • MLDA: The output of MLDA is a set of linear discriminants (LDs). These LDs are also linear combinations of the original features but are chosen to maximize class separation. Typically, the number of LDs is less than or equal to the number of classes minus one.

5. Use Cases: