ABSTRACT
Dual PCA is a variant of Classical Principal Component Analysis (PCA) which is used when the dimensionality of the data is more than the number of data points. Singular Value Decomposition (SVD) is used to solve the eigendecomposition for PCA, which works just perfectly when there are more data points than the dimensionality of the data. However, if it's the other way around, eigendecomposition using SVD would be computationally expensive. So, we use this variant of PCA. In this chapter, we discuss Dual PCA and some of the most commonly used cases. Furthermore, Dual PCA follows all the advantages and limitations of PCA.
