Principal Component Analysis (page 2 of 5)

PCA was invented in 1901 by Karl Pearson and again by Harold Hotelling in the 1930s. PCA is an orthogonal linear transformation of our data (features) to a new coordinate system such that the greatest variance of the data comes to lie on the first coordinate (called the first principal component), the second greatest variance on the second coordinate, and so on. Principal components are new features that are constructed as linear combinations of the original features such that the new features are uncorrelated. Transforming features to principal components allows us to reduce dimensionality without losing too much signal or information.