Supervised Principal Component Analysis
Description
Dimension reduction of complex data with supervision from auxiliary information. The package contains a series of methods for different data types (e.g., multi-view or multi-way data) including the supervised singular value decomposition (SupSVD), supervised sparse and functional principal component (SupSFPC), supervised integrated factor analysis (SIFA) and supervised PARAFAC/CANDECOMP factorization (SupCP). When auxiliary data are available and potentially affect the intrinsic structure of the data of interest, the methods will accurately recover the underlying low-rank structure by taking into account the supervision from the auxiliary data. For more details, see the paper by Gen Li, .