Search results for "partial least squares"
showing 10 items of 146 documents
Identification of compounds responsible for the odorant properties of aromatic caramel
2012
Aromatic caramel results from the heat treatment of sugars under specific temperature conditions. Because of its richness in aroma compounds and its pleasant organoleptic properties, caramel is widely used in the food industry. However, the composition of the volatile odorant fraction has not been completely elucidated. The aim of this work was thus to identify the volatile odorant compounds responsible for caramel sensory properties using a multivariate statistical technique. Four aromatic caramels differing in terms of their carbohydrate composition and cooking process were chosen. Odorant compounds were screened by gas chromatography–olfactometry (GC-O) and identified by GC–mass spectrom…
Determination of lecithin and soybean oil in dietary supplements using partial least squares-Fourier transform infrared spectroscopy.
2008
Lecithin and soybean oil in dietary supplements were determined by Fourier transform infrared spectrometry transmission measurements in dichloromethane in combination with a partial least squares (PLS) regression. Two different PLS models were developed, using 16 synthetic mixtures of analytes in dichloromethane, making measurements in the spectral range from 931.8 to 1252.3 cm(-1) for lecithin and from 911.4 to 1246.9 cm(-1) and 1695.3 to 1774.5 cm(-1) for soybean oil. Seven products from the Spanish market with lecithin concentrations between 21.1% and 99.1% and soybean oil concentrations between 0% and 37.2% were analyzed by the proposed method and the data was compared to a chromatograp…
Simultaneous Kinetic Determination of Carbamate Pesticides after Derivatization withp-Aminophenol by Using Partial Least Squares
1996
Abstract A method has been developed for the fast spectrophotometric determination of propoxur, carbaryl, and ethiofencarb in water samples using injection analysis in the stopped-flow mode. The method is based on the reaction between p -aminophenol and the phenolic compounds obtained from the pesticides, after a previous hydrolysis with 0.05 M NaOH at room temperature for 15 min. The partial least-squares treatment of the spectrophotometry kinetic data provides a simultaneous determination of the three carbamate pesticides assayed with a relative accuracy error lower than 5% in complex mixtures also containing formetanate, which is only partially hydrolyzed under the experimental condition…
Prediction of soil texture distributions using VNIR-SWIR reflectance spectroscopy
2013
Abstract Reflectance spectroscopy provides an alternate method to non-destructively characterize key soil properties. Different approaches, including chemometrics techniques or specific absorption features, have been proposed to estimate soil properties from visible and near-infrared (VNIR, 400-1200 nm) and shortwave infrared (SWIR, 1200-2500 nm) reflectance domains. The main goal of this study was to test the performance of two distinct methods for soil texture estimation by VNIR-SWIR reflectance measurements: i) the Continuum Removal (CR) technique that was used to correlate specific spectral absorption features with clay, silt and sand content, and ii) the Partial Least-Squares Regressio…
Missing Data
2009
In this chapter, we deal with the problem of missing data in principal component analysis (PCA) and partial least squares (PLS) methods. First, we review several statistical methods proposed in the literature for handling missing data. Both single and multiple imputation (MI) methods are studied and compared using simulated data. After this, we particularize the missing data problem for building and exploiting multivariate calibration models. Several approaches proposed in the literature are introduced and their performance compared based on several real data sets.
A Comparison of Advanced Regression Algorithms for Quantifying Urban Land Cover
2014
Quantitative methods for mapping sub-pixel land cover fractions are gaining increasing attention, particularly with regard to upcoming hyperspectral satellite missions. We evaluated five advanced regression algorithms combined with synthetically mixed training data for quantifying urban land cover from HyMap data at 3.6 and 9 m spatial resolution. Methods included support vector regression (SVR), kernel ridge regression (KRR), artificial neural networks (NN), random forest regression (RFR) and partial least squares regression (PLSR). Our experiments demonstrate that both kernel methods SVR and KRR yield high accuracies for mapping complex urban surface types, i.e., rooftops, pavements, gras…
Specifications of model development
2016
Chapter 4 goes into detail with the specifications of the model and model validation. Why Partial Least Squares (PLS), a structural equation modelling approach, is chosen as the method for model testing is explained in section 4.1, while 4.2 describes the survey conducted to collect data for model testing. Section 4.3 goes into detail with the PLS approach, its theoretical background and its application to the research question, before section 4.4 outlines the necessary operationalisation of the constructs introduced in chapter 3.
Fast-multivariate optimization of chiral separations in capillary electrophoresis: Anticipative strategies
2014
Abstract The design of experiments (DOE) is a good option for rationally limiting the number of experiments required to achieve the enantioresolution (Rs) of a chiral compound in capillary electrophoresis. In some cases, the modeled Rs after DOE analysis can be unsatisfactory, maybe because the range of the explored factors (DOE domain) was not the adequate. In these cases, anticipative strategies can be an alternative to the repetition of the process (e.g. a new DOE), to save time and money. In this work, multiple linear regression (MLR)-steepest ascent and a new anticipative strategy based on a multiple response-partial least squares model (called PLS2-prediction) are examined as post-DOE…
Comparative study of reflectance cells for PLS-FTIR determination of sugars in soft drinks
1998
A comparative study has been carried out between a horizontal Attenuated Total Reflectance (h-ATR) cell and a Cylindrical Internal Reflection (CIR) cell (CIRCLE cell). Both cells were employed for the determination of glucose, fructose, sucrose and total sugar in soft drinks and fruit juices using absorbance measurements at two resolution values (4 and 8 cm–1). Data were processed by Partial Least Squares regression (PLS). Using a resolution of 4 cm–1, mean relative errors of prediction from 0.6% to 1.3% were obtained using the horizontal-ATR and from 0.6% to 2.8% using the CIRCLE cell resulting in a limit of detection from 0.03% in the case of sucrose to 0.21% for total sugar for the h-ATR…
Analyser of chromium and/or cobalt
2003
Abstract Two stopped-flow manifolds have been proposed for individual or simultaneous determination of chromium and cobalt in water samples. Automated procedures based on multicommutation systems have emphasised the differences of their catalytic effect in luminol–hydrogen peroxide chemiluminescence reaction. A more rapid decay of signal was observed for Co for both configurations (flow injection or continuous injection). The influence of chemical and hydrodynamic variables has been studied in order to establish the robustness of method. The analysis rate was lower 1.5 min per replicate. Chemometric tools have been employed for the resolution of their contributions. Partial least squares (P…