PCA
In plain terms
A technique reducing dimensions by finding principal directions of variance.
Going deeper
Principal Component Analysis projects data onto the directions of greatest variance, compressing it while retaining most information. It finds the most informative axes.
Under the hood
Linear and unsupervised; components are orthogonal and ranked by explained variance.
Examples
Dimensionality reduction, visualization, noise reduction, feature compression.
Think of it like...
Like photographing a 3D object from the angle that reveals the most detail in 2D.