PR 05 · Does maximum variance preserve the classes?
IntermediatePR 05 · Feature Selection, Transforms and DimensionalityInteractive plot
Description
Rotate a one-dimensional projection and compare reconstruction error with class separation.
Try this: Find the projection that minimizes reconstruction error, then the one that separates the colored classes. Why are they different?
Hints
The chapter section that uses this lab:
Lab · PR 05 · Does maximum variance preserve the classes?
- Retained variance fraction
- 0.662
- Mean squared reconstruction distance
- 1.917
- Projected class-centroid gap
- 1
At 0°, both classes project to the same six positions. At 90°, within-class horizontal variation disappears and class centers remain distinct.
Try this: Find the projection that minimizes reconstruction error, then the one that separates the colored classes. Why are they different?
The centered synthetic observations have horizontal variation within each class and a vertical class difference. Principal component analysis maximizes retained variance without using labels. The centroid gap shown here is descriptive, not a complete classification metric.