Abstract:Selecting optimal
combinations of preprocessing methods is a major
holdup for chemometric analysis. The analyst decides which method(s)
to apply to the data, frequently by highly subjective or inefficient
means, such as user experience or trial and error. Here, we present
a user-friendly method using optimal experimental designs for selecting
preprocessing transformations. We applied this strategy to optimize
partial least square regression (PLSR) analysis of Stokes Raman spectra
to quantify hydroxylammonium (… Show more
“…S8 ‡). 29,30 The RMSEP% for Sm 3+ , HNO 3 , and temperature all fall below the goal level of 5%. While the U(VI) RMSEP% misses this objective, the ensemble model vastly improves the prediction accuracy compared to the initial PLSR models.…”
Section: Analyst Papermentioning
confidence: 92%
“…Numerous derivative orders, polynomial orders, and smoothing points were tested. 30 Trimming the spectra reduces the dimensionality of large spectra files and can help reduce the error of prediction in multivariate modeling. The trimmed region included in the chemometric model was varied based on the behavior being modeled.…”
Section: Multivariate Analysis and Preprocessingmentioning
confidence: 99%
“…A Tukey-Kramer significance test was used to statistically compare the RMSEP values for multiple regression models following a method outlined previously. 29,30 Additional details can be found in the ESI. ‡…”
Section: Statistical Comparisonmentioning
confidence: 99%
“…In addition to U(VI) and Sm 3+ emission peaks, Stokes Raman scattering features corresponding to free acid (H + ), nitrate anions (NO 3 − ), and the O-H stretching region were identified simultaneously. 26,30 Unperturbed nitrate ions have three Raman active bands including ν 1 (∼1048 cm −1 ), ν 3 (∼1415 cm −1 ), and ν 4 (∼717 cm −1 ). The O-H vibrational stretching region consists of several overlapping bands attributed to various H 2 O and O-H (free and bound) vibrations (455-475 nm).…”
Section: Analyst Papermentioning
confidence: 99%
“…These samples can be selected using optimal experimental designs that are the most flexible and effective option when a small number of experimental runs is desired. [29][30][31][32][33] Although PLSR has been used with great success to model systems with overlapping spectral features in numerous systems, 27 it does not always account for systems with a high degree of multicollinearity. 34,35 Multicollinearity must be accounted for in a LIFS system for monitoring U(VI) concentration because two independent variables (i.e., U(VI) concentration and temperature) are highly correlated in the regression model.…”
Laser-induced fluorescence spectroscopy (LIFS), Raman spectroscopy, and a stacked regression ensemble was developed for near real-time quantification of uranium (VI) (1–100 µg∙mL-1), samarium (0–200 µg∙mL-1) and nitric acid (0.1–4 M)...
“…S8 ‡). 29,30 The RMSEP% for Sm 3+ , HNO 3 , and temperature all fall below the goal level of 5%. While the U(VI) RMSEP% misses this objective, the ensemble model vastly improves the prediction accuracy compared to the initial PLSR models.…”
Section: Analyst Papermentioning
confidence: 92%
“…Numerous derivative orders, polynomial orders, and smoothing points were tested. 30 Trimming the spectra reduces the dimensionality of large spectra files and can help reduce the error of prediction in multivariate modeling. The trimmed region included in the chemometric model was varied based on the behavior being modeled.…”
Section: Multivariate Analysis and Preprocessingmentioning
confidence: 99%
“…A Tukey-Kramer significance test was used to statistically compare the RMSEP values for multiple regression models following a method outlined previously. 29,30 Additional details can be found in the ESI. ‡…”
Section: Statistical Comparisonmentioning
confidence: 99%
“…In addition to U(VI) and Sm 3+ emission peaks, Stokes Raman scattering features corresponding to free acid (H + ), nitrate anions (NO 3 − ), and the O-H stretching region were identified simultaneously. 26,30 Unperturbed nitrate ions have three Raman active bands including ν 1 (∼1048 cm −1 ), ν 3 (∼1415 cm −1 ), and ν 4 (∼717 cm −1 ). The O-H vibrational stretching region consists of several overlapping bands attributed to various H 2 O and O-H (free and bound) vibrations (455-475 nm).…”
Section: Analyst Papermentioning
confidence: 99%
“…These samples can be selected using optimal experimental designs that are the most flexible and effective option when a small number of experimental runs is desired. [29][30][31][32][33] Although PLSR has been used with great success to model systems with overlapping spectral features in numerous systems, 27 it does not always account for systems with a high degree of multicollinearity. 34,35 Multicollinearity must be accounted for in a LIFS system for monitoring U(VI) concentration because two independent variables (i.e., U(VI) concentration and temperature) are highly correlated in the regression model.…”
Laser-induced fluorescence spectroscopy (LIFS), Raman spectroscopy, and a stacked regression ensemble was developed for near real-time quantification of uranium (VI) (1–100 µg∙mL-1), samarium (0–200 µg∙mL-1) and nitric acid (0.1–4 M)...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.