The successive projections algorithm for spectral variable selection in classification problems

Pontes, Márcio José Coelho; Galvão, Roberto Kawakami Harrop; Araújo, Mário César Ugulino de; Moreira, Pablo N.T.; Neto, Osmundo D. Pessoa; José, Gledson Emídio; Saldanha, Teresa Cristina Bezerra

doi:10.1016/j.chemolab.2004.12.001

Cited by 155 publications

(62 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…SPA has been compared to the genetic algorithm, which is a popular method for variable selection in multivariate calibration. The results were proven to be in favor of SPA , Pontes et al, 2005. It was found to be less sensitive to instrumental noise than the genetic algorithm SPA-MLR models, which have been proven to be comparable to or better than the full spectrum PLS or PCR models for UV-Vis (Ugolino and NIR spectra analysis ).…”

Section: Discussionmentioning

confidence: 64%

“…Models with 10 variables had lower or equal classification errors when compared to those with 6 variables. This suggests that an additional variable added to 10-variable models may improve their discriminatory power although Pontes et al (2005) show that this discriminatory power may be improved only to a certain extent. At some point, adding an additional variable would cause the deterioration of the model.…”

Section: Synchronous Fluorescence Spectra Of Olive Oil Samplesmentioning

confidence: 94%

“…It was found to be less sensitive to instrumental noise than the genetic algorithm SPA-MLR models, which have been proven to be comparable to or better than the full spectrum PLS or PCR models for UV-Vis (Ugolino and NIR spectra analysis ). Promising results were obtained for SPA-LDA as a method for dealing with low-resolution UV-Vis spectra with strong overlapping (Pontes et al, 2005). Recently, SPA was applied for variable selection to determine the organic acid of plum vinegar {Vis/NIR} (Liu and He, 2009), classification of coffees {UV-Vis} (Polari Souto et al, 2010), and edible oils {UV-Vis} (Pontes et al, 2005).…”

Section: Discussionmentioning

confidence: 99%

See 2 more Smart Citations

Discrimination of edible olive oils by means of synchronous fluorescence spectroscopy with multivariate data analysis

Dankowska¹,

Małecka²,

Kowalewski³

2013

Grasas y Aceites

View full text Add to dashboard Cite

RESUMEN Discriminación de aceites de oliva comestibles mediante espectroscopía de fluorescencia sincrónica y análisis multivarianteSe ha investigado el potencial de la espectroscopía de fluorescencia para la clasificación de los aceite de oliva. Para ello, se recogieron espectros de fluorescencia sincrónica en el rango de 240-700 nm con intervalos de longitud de onda de 10, 30, 60 y 80 nm. Las proyecciones algorítmicas sucesivas (SPA) se aplicaron para la determinación de las longitudes de onda representativas mientras que el método de análisis discriminante lineal (LDA) se empleó para clasificar los aceites de oliva. Se obtuvo un error de clasificación del método bajo (0,9-6,4%) para las medidas recogidas en todos los intervalos de onda. La mejor precisión de clasificación se obtuvo para intensidades de fluorescencia sincrónica adquiridos a 10 longitudes de onda seleccionadas con intervalos de longitud de onda de 10 nm. PALABRAS CLAVE: Aceite de oliva refinado -Aceite de oliva virgen extra -Aceite de orujo -Espectroscopía de fluorescencia sincrónica -Proyecciones algorítmicas sucesivas. SUMMARY Discrimination of edible olive oils by means of synchronous fluorescence spectroscopy with multivariate data analysisThe potential of fluorescence spectroscopy for the classification of olive oils was investigated. Synchronous fluorescence spectra were collected in the region of 240-700 nm with the wavelength intervals of 10, 30, 60 and 80 nm. Successive projection algorithm (SPA) was applied for the determination of representative wavelengths while the linear discriminant analysis (LDA) method was used to classify olive oils. The classification error of the method was low (0,9-6,4%) for measurements collected at all wavelength intervals. The best classification accuracy was obtained for synchronous fluorescence intensities acquired at 10 selected wavelengths with the wavelength interval equal to 10 nm. KEY-WORDS: Extra virgin olive oil -Pomace olive oil -Refined olive oil -Successive projections algorithmSynchronous fluorescence spectroscopy.

show abstract

Section: Discussionmentioning

confidence: 64%

Section: Synchronous Fluorescence Spectra Of Olive Oil Samplesmentioning

confidence: 94%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Discrimination of edible olive oils by means of synchronous fluorescence spectroscopy with multivariate data analysis

Dankowska¹,

Małecka²,

Kowalewski³

2013

Grasas y Aceites

View full text Add to dashboard Cite

show abstract

“…UVE is one of the most prevalent variable selection methods that is widely used in analytical chemistry [25]; it is able to remove the variables that are not more informative than noise for modeling, and thus increase the model's predictive accuracy [26]. SPA is also a common method to select variables in multivariate modeling, and has been more favorable than the genetic algorithm [27]. SPA is an iterative forward selection method that adopts projection operations to choose variables of collinearity minimum.…”

Section: Optimal Wavelength Extraction and Multispectral Model Develomentioning

confidence: 99%

Evaluation of Near-Infrared Hyperspectral Imaging for Detection of Peanut and Walnut Powders in Whole Wheat Flour

Zhao

Wang

et al. 2018

Applied Sciences

View full text Add to dashboard Cite

Featured Application: Near-infrared hyperspectral imaging was employed for inspection of contaminant of peanut or walnut powders in whole wheat flour. A multispectral prediction model and visualization maps demonstrated the potential of the technique for the application of rapid quality control. Abstract:The general utilization of processing equipment in industry has increased the risk of foreign material contamination. For example, peanut and walnut contaminants in whole wheat flour, which typically a healthy food, are a threat to people who are allergic to nuts. The feasibility of utilizing near-infrared hyperspectral imaging to inspect peanut and walnut powder in whole wheat flour was evaluated herein. Hyperspectral images at wavelengths 950-1700 nm were acquired. A standard normal variate combined with the Savitzky-Golay first derivative spectral transformation was adopted for the development of a partial least squares regression (PLSR) model to predict contamination concentrations. A successive projection algorithm (SPA) and uninformative variable elimination (UVE) for feature wavelength selection were compared. Two individual prediction models for peanut or walnut-contaminated flour, and a general multispectral model for both peanut-contaminated flour and walnut-contaminated flour, were developed. The optimal general multispectral model had promising results, with a determination coefficient of prediction (R p 2 ) of 0.987, and a root mean square error of prediction (RMSEP) of 0.373%. Visualization maps based on multispectral PLSR models reflected the contamination concentration variations in a spatial manner. The results demonstrated that near-infrared hyperspectral imaging has the potential to inspect peanut and walnut powders in flour for rapid quality control.

show abstract

“…NIR analysis of sugar cane juice has utilized partial least squares (PLS) pruning for variable selection [5]. UV-VIS and NIR spectrometry of oils takes advantage of the successive projections algorithm (SPA) in large-scale variable selection [6].…”

Section: Introductionmentioning

confidence: 99%

Variable Selection and Grouping in a Paper Machine Application

Ahola

Juuso

Leiviskä

2007

INT J COMPUT COMMUN

View full text Add to dashboard Cite

This paper describes the possibilities of variable selection in large-scale industrial systems. It introduces knowledge-based, data-based and model-based methods for this purpose. As an example, Case-Based Reasoning application for the evaluation of the web break sensitivity in a paper machine is introduced. The application uses Linguistic Equations approach and basic Fuzzy Logic. The indicator combines the information of on-line measurements with expert knowledge and provides a continuous indication of the break sensitivity. The web break sensitivity defines the current operating situation at the paper mill and gives new information to the operators. Together with information of the most important variables this prediction gives operators enough time to react to the changing operating situation.

show abstract

The successive projections algorithm for spectral variable selection in classification problems

Cited by 155 publications

References 23 publications

Discrimination of edible olive oils by means of synchronous fluorescence spectroscopy with multivariate data analysis

Discrimination of edible olive oils by means of synchronous fluorescence spectroscopy with multivariate data analysis

Evaluation of Near-Infrared Hyperspectral Imaging for Detection of Peanut and Walnut Powders in Whole Wheat Flour

Variable Selection and Grouping in a Paper Machine Application

Contact Info

Product

Resources

About