Showing papers in "Chemometrics and Intelligent Laboratory Systems in 2001"

TL;DR: PLS-regression (PLSR) as mentioned in this paper is the PLS approach in its simplest, and in chemistry and technology, most used form (two-block predictive PLS) is a method for relating two data matrices, X and Y, by a linear multivariate model.

TL;DR: In this paper, the authors proposed a variable selection strategy for multivariate calibration using the Successive Projections Algorithm, a forward selection method which uses simple operations in a vector space to minimize variable collinearity.

TL;DR: The Monte Carlo cross validation developed in this paper is an asymptotically consistent method in determining the number of components in calibration model and can avoid an unnecessary large model and therefore decreases the risk of over-fitting for the calibration model.

TL;DR: The original chemometrics partial least squares (PLS) model with two blocks of variables linearly related to each other, has had several enhancements/extensions since the beginning of 19...

TL;DR: If the complete orthogonality constraint is loosened, the method improves considerably and simplifies the calibration model for the prediction of Y and the number of OSC algorithms that have appeared are compared from a theoretical viewpoint.

TL;DR: In this article, the authors present some useful methods for introductory analysis of variables and subsets in relation to PLS regression, and also present an approach to orthogonal scatter correction.

TL;DR: In this article, a density-based unsupervised clustering approach for detecting natural patterns in data (further denoted as NP) is presented, and its performance is illustrated for data sets with different types of clusters.

TL;DR: In this article, three major methods of experimental design, such as factorial design, orthogonal design, D-optimal design and uniform design, and their applications in chemistry and chemical engineering are reviewed and compared.

TL;DR: In this article, a survey of partial least squares regression with one y variable from a theoretical point of view is given, where general comments are made on the motivation as seen by a statistician to study particular chemometric methods, and the concept of soft modelling is criticized from the same angle.

TL;DR: The main concepts of the maximum likelihood approach in dealing with missing data are introduced and simple numerical examples of the application of ML are presented Differences between ML and other techniques of treating missing data.

TL;DR: The paper outlines the author's frustrating experiences in the 70'ies with two conflicting and equally over-ambitious and over-simplified modelling cultures - in traditional chemistry and in traditional statistics.

TL;DR: In this article, a stepwise variable selection based on the maximum Q cum 2 criterion, similar to the Stone-Geisser index, was proposed, depending on the number of eliminated variables.

TL;DR: In this article, a comparison between orthogonal signal correction (OSC) and net analyte signal (NAS) calculations is presented, and it is shown that the latter can be used as a preprocessing method comparable to the former, before the application of partial least-squares (PLS) to the filtered data.

TL;DR: In this article, three methods for multiway regression are compared: unfold partial least squares (PLS), multilinear PLS and multiway covariates regression (MCovR).

TL;DR: In this article, the partial least squares (PLS) Path Modeling approach is applied to a study of the cosmetic habits of women in the Ile-de-France region.

TL;DR: In this paper, the authors present MFA in detail and PLS path modelling more briefly, and also mention some links between MFA, PLS Path Modeling and PLS regression.

TL;DR: Martens et al. as discussed by the authors proposed a method for estimating the reliability of the cross-validated prediction error RMSEP, which is based on the data at hand and with little need for abstract distribution theory.

TL;DR: GA-optimized models exhibited about 20% improvement over the experimentally chosen best models, and improvement of more than 30% was achieved with respect to statistical response surface models.

TL;DR: This paper uses simulated first- and second-order reaction data as well as real data to characterize the performance of the soft modeling technique, which is able to resolve overlapped retention and spectral profiles and predict the rate constants for the reaction.

TL;DR: In this paper, a near-infrared set-up based on simultaneous detection of four wavelengths was applied for in-line moisture measurement during fluid bed granulation, and the back-propagation neural network approach was found to have most predictive power with the independent test data.

TL;DR: The aim is to improve the secondary aspect of PLS: the modeling of the independent variables, which is of importance, for example, in process monitoring, outlier detection and also, implicitly, for jackknifing of model parameters.

TL;DR: A novel algorithm based on the WPT for pattern recognition of signals, which operates both feature selection and classification at the same time: Wavelet Packet Transform for Efficient pattern Recognition of signals (WPTER).

TL;DR: An algorithm is proposed for compressing calibration spectra based on a wavelet transformation before performing the SVD, which could be accelerated up to a factor of 52 and concentrations of synthetic spectra are evaluated by means of the PCs obtained by the different PCA algorithms.

TL;DR: Univ Nice, Equipe Chimiometrie & Modelisat, Lab Aromes Syntheses & Interact, F-06108 Nice 2, France

TL;DR: This study investigated whether useful information can be extracted from an electroencephalographic data set with a very high number of modes and to determine which model is the most appropriate for this purpose, and found that Tucker 3 was the most suited model.

TL;DR: In this article, the effects of a direct orthogonalization before applying PCR and PLS are studied for several data sets and it is shown that the quality of the calibration model is usually not better when using DO, nor does the predictive quality change significantly in most cases.

TL;DR: In this paper, an extension of the linear Partial Least Squares (PLS) model to the nonlinear additive through the transformation of predictors by polynomial spline functions is presented.