Component Analysis and Discriminants (0) - PCA
Introduction In this article, I survey most of the component-analysis-type(or dimension reduction) of learning methods such as PCA(Principal Component Analysis), LDA(Linear Discriminant Analysis), FLD(Fischer Linear Discriminant), MDA(Multivariate Discriminant Analysis), CCA(Canonical Correlation Analysis). The following description is in chronological order. Whenever we move to the next methodology, the conditions of problems that we face are more complicated. There are so many articles about these methodologies. But I could not found any report that focuses on the relationship between methods. I need to know why our predecessors developed these: from PCA to CCA or other more complicated methods. My final goal is to set up a unified table of all methods using the same mathematical notation for easier understanding. PCA(Principal Component Analysis) Line Fitting on Sample Points Before we start to discuss PCA, let's review the ...