In this paper we present a comparison of three projection methods that use the eigenvectors of a matrix to investigate high-dimensional dataset: principal component analysis (PCA), principal component analysis followed by independent component analysis (PCA+ICA), and Laplacian projections. We demonstrate the application of these methods to a sponsored links search listings dataset and provide a...

We study sparse principal components analysis in the high-dimensional setting, where p (the number of variables) can be much larger than n (the number of observations). We prove optimal, non-asymptotic lower and upper bounds on the minimax estimation error for the leading eigenvector when it belongs to an lq ball for q ∈ [0, 1]. Our bounds are sharp in p and n for all q ∈ [0, 1] over a wide cla...

Face recognition is one of the Biometric characteristics for person identification. In this paper, Face recognition is done using two feature extraction techniques PCA (Principal Component Analysis) and MPCA (Modular Principal Component Analysis). PCA is a linear projection method in which dimensionality reduction is applied to the original image space. MPCA is an improved version of PCA in whi...

eight selected heavy metals (cu, zn, pb, cd, cr, mn, fe and al) in surface and sub-surfacesediments in the northern coast of sfax (tunisia) were studied in order to assess the sediment quality and tohighlight the anthropogenic contributions to heavy metal distributions in the two study sediment levels.multiple chemometric approaches based on geographic information system (gis), enrichment facto...

the species of dioscorea (yam) are regarded as a staple food crop for millions of people in the tropical and subtropical regions of the world. it is regarded as an important food crop next to cereals and grains due to high yield storage of carbohydrates. economically, only few species are recognized for cultivation from agricultural point of view, in spite of its large species diversity. the sp...

In this letter we present a coupled Helmholtz machine for principal component analysis (PCA), where sub-machines are related through sharing some latent variables and associated weights. Then, we present a wake-sleep PCA algorithm for training the coupled Helmholtz machine, showing that the algorithm iteratively determines principal eigenvectors of a data covariance matrix without any rotationa...