PCA shrinks data into a few new features with the biggest patterns.
PCA is like squinting at a messy school photo. You miss the braces, but you still spot the class clown.
It turns many columns into a few new ones for simple charts or noise cleanup. Squeeze too hard, and small details vanish.
Dim. Reduction
PCA uses a few main parts to sum up the data.
Feature Selection
PCA mixes old features into new ones instead of picking them.
Unsupervised Learning
PCA needs no labels and finds the biggest changes in the data.
Autoencoder
PCA and an Autoencoder both compress data, but an Autoencoder is often more flexible.