9.4 Dimensionality Reduction

Dimensionality reduction takes data with many features (dimensions) and represents it with far fewer, while preserving as much of the meaningful information as possible. Principal Component Analysis (PCA) is the classic method: it finds new combined axes that capture the most variation in the data.

Imagine describing a person with 50 measurements. Many overlap — height and weight, for instance, move together. Dimensionality reduction might compress those 50 numbers into 3 that still capture most of what distinguishes people, making the data easier to plot and faster to process.