ABSTRACT
Principal component analysis (PCA) is a statistical technique of representing high-dimensional data in a low-dimensional space. PCA is usually used to reduce the dimensionality of data so that the data can be further visualized or analyzed in a low-dimensional space. This chapter reviews multivariate statistics and matrix algebra is first given to lay the mathematical foundation of PCA. It describes a list of software packages that support PCA. Using a few principal components to represent the data, the data can be further visualized in a one-, two-, or three-dimensional space of the principal components to observe data patterns, or can be mined or analyzed to uncover data patterns of principal components. Note that the mathematical meaning of each principal component as the linear combination of the original data variable does not necessarily have a meaningful interpretation in the problem domain.
