Optimization of Partial-Least-Squares Calibration Models by Simulation of Instrumental Perturbations

1997 ◽  
Vol 69 (16) ◽  
pp. 3391-3399 ◽  
Author(s):  
Frédéric Despagne ◽  
Désiré-Luc Massart ◽  
Onno E. de Noord
Keyword(s):  
Least Squares ◽  
Partial Least Squares ◽  
Calibration Models

Estimation of Viscosity of Biodiesel-Diesel Blends with near Infrared Spectroscopy

2012 ◽  
Vol 608-609 ◽  
pp. 324-327
Author(s):  
Wei Bo Zhang ◽  
Ming Ming Wu
Keyword(s):  
Infrared Spectroscopy ◽  
Least Squares ◽  
Partial Least Squares ◽  
Near Infrared Spectroscopy ◽  
Kinematic Viscosity ◽  
Near Infrared ◽  
Diesel Oil ◽  
Processing Methods ◽  
Calibration Models ◽  
Using Data

Biodiesel is one of the most important substitutes for diesel oil. This work reports the use of near Infrared Spectroscopy (NIR) to estimate the kinematic viscosity value of biodiesel-diesel blends. Partial least squares models were developed using data of different spectra regions and different pre-processing methods were employed for developing the calibration models. The results indicate that NIR can be used in biodiesel-diesel blends properties detecting.

scholarly journals Tea and Chicory Extract Characterization, Classification and Authentication by Non-Targeted HPLC-UV-FLD Fingerprinting and Chemometrics

Foods ◽  
2021 ◽  
Vol 10 (12) ◽  
pp. 2935
Author(s):  
Josep Pons ◽  
Àlex Bedmar ◽  
Nerea Núñez ◽  
Javier Saurina ◽  
Oscar Núñez
Keyword(s):  
Least Squares ◽  
Partial Least Squares ◽  
External Validation ◽  
Prediction Errors ◽  
Classification Rate ◽  
White Tea ◽  
Oolong Tea ◽  
Financial Benefits ◽  
Calibration Models ◽  
Almost All

Tea is a widely consumed drink in the world which is susceptible to undergoing adulterations to reduce manufacturing costs and rise financial benefits. The development of simple analytical methodologies to assess tea authenticity, as well as to detect and quantify frauds, is an important matter considering the rise of adulteration issues in recent years. In the present study, untargeted HPLC-UV and HPLC-FLD fingerprinting methods were employed to characterize, classify, and authenticate tea extracts belonging to different varieties (red, green, black, oolong, and white teas) by partial least squares-discriminant analysis (PLS-DA), as well as to detect and quantify adulteration frauds when chicory was used as the adulterant by partial least squares (PLS) regression, to ensure the authenticity and integrity of foodstuffs. Overall, PLS-DA showed a good classification and grouping of the tea samples according to the tea variety and, except for some white tea extracts, perfectly discriminated from the chicory ones. One hundred percent classification rates for the PLS-DA calibration models were achieved, except for green and oolong tea when HPLC-FLD fingerprints were employed, which showed classification rates of 96.43% and 95.45%, respectively. Good predictions were also accomplished, also showing, in almost all the cases, a 100% classification rate for prediction, with the exception of white tea and oolong tea when HPLC-UV fingerprints were employed that exhibited a classification rate of 77.78% and 88.89%, respectively. Good PLS results for chicory adulteration detection and quantitation were also accomplished, with calibration, cross-validation, and external validation errors beneath 1.4%, 6.4%, and 3.7%, respectively. Acceptable prediction errors (below 21.7%) were also observed, except for white tea extracts that showed higher errors which were attributed to the low sample variability available.

scholarly journals Can We Trust Score Plots?

Metabolites ◽  
2020 ◽  
Vol 10 (7) ◽  
pp. 278 ◽  
Author(s):  
Marta Bevilacqua ◽  
Rasmus Bro
Keyword(s):  
Discriminant Analysis ◽  
Least Squares ◽  
Partial Least Squares ◽  
Partial Least Squares Regression ◽  
Classification Models ◽  
Least Squares Regression ◽  
Calibration Models ◽  
Accepted Practice ◽  
Component Models

In this paper, we discuss the validity of using score plots of component models such as partial least squares regression, especially when these models are used for building classification models, and models derived from partial least squares regression for discriminant analysis (PLS-DA). Using examples and simulations, it is shown that the currently accepted practice of showing score plots from calibration models may give misleading interpretations. It is suggested and shown that the problem can be solved by replacing the currently used calibrated score plots with cross-validated score plots.

Application of Natural Frequencies for Prediction of Apple Texture Based on Partial Least Squares Regression

2017 ◽  
Vol 13 (10) ◽  
Author(s):  
Jumin Hou ◽  
Yonghai Sun ◽  
Fangyuan Chen ◽  
Lu Wang ◽  
Xue Bai ◽  
...  
Keyword(s):  
Least Squares ◽  
Partial Least Squares ◽  
Partial Least Squares Regression ◽  
Natural Frequencies ◽  
Prediction Models ◽  
Good Prediction ◽  
Least Squares Regression ◽  
Prediction Ability ◽  
Initial Modulus ◽  
Calibration Models

AbstractExperimental modal analysis was performed to identify natural frequencies to predict the texture of inhomogeneous tissues of apple (Malus domestinacv. ‘Golden Delicious’). Partial least squares calibration models based on natural frequencies with or without weight and density were created for predicting apple texture representing by yield gradient and initial modulus. The prediction models shown good prediction ability for texture of skin but impossible for flesh (all determination coefficients for skin models were more than 0.5 while for flesh models less than 0.5). A nondestructive and rapid method was provided to evaluate the fruit texture.

scholarly journals Fuel park wastewater monitoring with UV-Vis spectra and partial least squares models

2008 ◽  
Vol 27 (1) ◽  
pp. 19
Author(s):  
Nídia D. Lourenço ◽  
José C. Menezes ◽  
Helena M. Pinheiro ◽  
David Diniz
Keyword(s):  
Least Squares ◽  
Partial Least Squares ◽  
Mean Squared Error ◽  
Treatment Plant ◽  
Oxygen Demand ◽  
Original Dataset ◽  
Calibration Models ◽  
Wastewater Quality ◽  
On Line ◽  
Determination Of Parameters

In the context of the high application potentials for on-line and in-line measurements in wastewater quality monitoring, UV-Vis spectrophotometry has received recent attention. In the present work the development of Partial Least Squares (PLS) calibration models for the fast determination of parameters legally required at the discharge of a fuel park wastewater treatment plant was attempted using UV-Vis spectra and the corresponding standard analytical values. A PLS model was developed and validated for Chemical Oxygen Demand (COD) determination with 3.8 mg O2/l RMSEP (root mean squared error of prediction). The models developed for 5-day Biochemical Oxygen Demand (BOD5) and Total Suspended Solids (TSS) were validated (3.5 mg O2/l and 2.9 mg/l RMSEP, respectively), but presented some predictive limitations mainly attributed to the narrow interval of concentration values present in the original dataset.

Spectral Variable Selection for Partial Least Squares Calibration Applied to Authentication and Quantification of Extra Virgin Olive Oils Using Fourier Transform Raman Spectroscopy

2005 ◽  
Vol 59 (10) ◽  
pp. 1286-1294 ◽  
Author(s):  
H. Michael Heise ◽  
Uwe Damm ◽  
Peter Lampen ◽  
Antony N. Davies ◽  
Peter S. McIntyre
Keyword(s):  
Fourier Transform ◽  
Variable Selection ◽  
Olive Oil ◽  
Least Squares ◽  
Partial Least Squares ◽  
Sunflower Oil ◽  
Cross Validation ◽  
Calibration Models ◽  
Fourier Transform Raman Spectroscopy ◽  
Spectral Variable

The limits of quantitative multivariate assays for the analysis of extra virgin olive oil samples from various Greek sites adulterated by sunflower oil have been evaluated based on their Fourier transform (FT) Raman spectra. Different strategies for wavelength selection were tested for calculating optimal partial least squares (PLS) models. Compared to the full spectrum methods previously applied, the optimum standard error of prediction (SEP) for the sunflower oil concentrations in spiked olive oil samples could be significantly reduced. One efficient approach (PMMS, pair-wise minima and maxima selection) used a special variable selection strategy based on a pair-wise consideration of significant respective minima and maxima of PLS regression vectors, calculated for broad spectral intervals and a low number of PLS factors. PMMS provided robust calibration models with a small number of variables. On the other hand, the Tabu search strategy recently published (search process guided by restrictions leading to Tabu list) achieved lower SEP values but at the cost of extensive computing time when searching for a global minimum and less robust calibration models. Robustness was tested by using packages of ten and twenty randomly selected samples within cross-validation for calculating independent prediction values. The best SEP values for a one year's harvest with a total number of 66 Cretian samples were obtained by such spectral variable optimized PLS calibration models using leave-20-out cross-validation (values between 0.5 and 0.7% by weight). For the more complex population of olive oil samples from all over Greece (total number of 92 samples), results were between 0.7 and 0.9% by weight with a cross-validation sample package size of 20. Notably, the calibration method with Tabu variable selection has been shown to be a valid chemometric approach by which a single model can be applied with a low SEP of 1.4% for olive oil samples across three different harvest years.

Sign in / Sign up

Export Citation Format

Share Document