Nonlinear Principal Component Analysis and Its Applications

Shows that PCA, nonlinear PCA, and MCA can be integrated as a single formulation, which can easily be extended to several applicationsProvides an acceleration algorithm that speeds up the convergent sequences generated by the alternating least squares and is a remedy for computational costIntroduces applications related to nonlinear PCA: variable selection for mixed measurement levels data, sparse multiple correspondence analysis, and joint dimension reduction and clusteringIncludes supplementary material: sn.pub/extras