Comprehensive new approaches for variable selection using ordered predictors selection

Roque, Jussara V.; Cardoso, Wilson; Peternelli, Luíz Alexandre; Teófilo, Reinaldo F.

doi:10.1016/j.aca.2019.05.039

Cited by 40 publications

(28 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The KS algorithm is often used in the area of chemometrics (HONORATO et al, 2007;ROQUE et al, 2019;SOUSA et al, 2011;WU et al, 1996) for presenting good results, and for being easy to be applied and understood. The RS method, also simple and easy to understand, is popular in statistical applications in several areas (AKDEMIR et al, 2015;HE,.…”

Section: Crop Productionmentioning

confidence: 99%

See 1 more Smart Citation

Kennard-Stone method outperforms the Random Sampling in the selection of calibration samples in SNPs and NIR data

2022

View full text Add to dashboard Cite

Splitting the whole dataset into training and testing subsets is a crucial part of optimizing models. This study evaluated the influence of the choice of the training subset in the construction of predictive models, as well as on their validation. For this purpose we assessed the Kennard-Stone (KS) and the Random Sampling (RS) methods in near-infrared spectroscopy data (NIR) and marker data SNPs (Single Nucleotide Polymorphisms). It is worth noting that in SNPs data, there is no knowledge of reports in the literature regarding the use of the KS method. For the construction and validation of the models, the partial least squares (PLS) estimation method and the Bayesian Lasso (BLASSO) proved to be more efficient for NIR data and for marker data SNPs, respectively. The evaluation of the predictive capacity of the models obtained after the data partition occurred through the correlation between the predicted and the observed values, and the corresponding square root of the mean squared error of prediction. For both datasets, results indicated that the results from KS and RS methods differ statistically from each other by the F test (P-value < 0.01). The KS method showed to be more efficient than RS in practically all repetitions. Also, KS method has the advantage of being easy and fast to be applied and also to select the same samples, which provides excellent benefits in the following analyses.

show abstract

Section: Crop Productionmentioning

confidence: 99%

“…In the NIR data approach, the KS method is widely used for data partitioning (GOGÉ et al, 2012;NASCIMENTO et al, 2016;ROQUE et al, 2019). Some studies affirmed the use of the KS method for partitioning data sets compared to other methods (SAPTORO et al, 2012;SIANO;GOICOECHEA, 2007;WU et al, 1996).…”

Section: Nir Datasetmentioning

confidence: 99%

Kennard-Stone method outperforms the Random Sampling in the selection of calibration samples in SNPs and NIR data

2022

View full text Add to dashboard Cite

show abstract

“…Details about these vector calculations can be found elsewhere. 4,5 The differentiated variables are sorted in descending order. First, a subset of 20 variables (window) is selected to build and evaluate the first model.…”

Section: Methodsmentioning

confidence: 99%

“…Finally, the evaluated variable sets (initial window and its extensions) are compared using the quality parameters calculated during validations. 4,5 In this study, stalk samples were classified according to a range of FC and PP determined by reference analysis. Samples were defined in two classes for FC and PP models.…”

Section: Methodsmentioning

confidence: 99%

“…This work aims to present a new method for the early selection of sugarcane genotypes by determination of FC and PP using NIR combined with partial least squares discriminant analysis (PLS-DA) and a method of variable selection, the ordered predictors selection (OPS). 4,5 NIR spectra were obtained directly on the stalk, which is the main innovation of this work. Reflectance NIR spectra were collected using a Fourier transform near infrared (FT-NIR) spectrometer (Thermo Scientific Antaris II) controlled with TQ Analysis software.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Phenotypic classification of sugarcane from near infrared spectra obtained directly from stalk using ordered predictors selection and partial least squares-discriminant analysis

Peternelli

Barbosa

Roque

et al. 2019

Proceedings of the 18th International Conference on Near Infrared Spectroscopy

View full text Add to dashboard Cite

A new method was developed for the early selection of sugarcane genotypes using near infrared spectroscopy combined with partial least squares discriminant analysis (PLS-DA) and a variable selection method named ordered predictors selection (OPS). The use of the OPS method improved the predictive capacity of PLS-DA models to classify the sugarcane samples correctly according to fiber content (FC) and pol percent (PP).

show abstract

Mapping the Distribution of Additives Within Polymer Films Through Near-Infrared Spectroscopy and Hyperspectral Imaging

Roque,

Pola,

Terra

et al. 2024

Methods and Protocols in Food Science

View full text Add to dashboard Cite

Comprehensive new approaches for variable selection using ordered predictors selection

Cited by 40 publications

References 34 publications

Kennard-Stone method outperforms the Random Sampling in the selection of calibration samples in SNPs and NIR data

Kennard-Stone method outperforms the Random Sampling in the selection of calibration samples in SNPs and NIR data

Phenotypic classification of sugarcane from near infrared spectra obtained directly from stalk using ordered predictors selection and partial least squares-discriminant analysis

Mapping the Distribution of Additives Within Polymer Films Through Near-Infrared Spectroscopy and Hyperspectral Imaging

Contact Info

Product

Resources

About