Novel feature selection method based on Stochastic Methods Coupled to Support Vector Machines using H- NMR data (data of olive and hazelnut oils)

One of the principal inconveniences that analysis and information processing presents is that of the representation of dataset. Normally, one encounters a high number of samples, each one with thousands of variables, and in many cases with irrelevant information and noise. Therefore,...

Full description

Bibliographic Details
Main Authors:	Oscar Eduardo Gualdron, Claudia Isaza, Cristhian Manuel Duran
Format:	Article
Language:	English
Published:	Universidad Industrial de Santander 2014-12-01
Series:	Revista Ion
Subjects:	eature selection H-NMR simulated annealing support vector machine olive oil hazelnut oil
Online Access:	http://revistas.uis.edu.co/index.php/revistaion/article/view/4594/4814

id	doaj-6f438c010d924a53b75eb52c7f87e1ca
record_format	Article
spelling	doaj-6f438c010d924a53b75eb52c7f87e1ca2020-11-25T01:08:02ZengUniversidad Industrial de SantanderRevista Ion0120-100X2145-84802014-12-012721728Novel feature selection method based on Stochastic Methods Coupled to Support Vector Machines using H- NMR data (data of olive and hazelnut oils)Oscar Eduardo Gualdron0Claudia Isaza1Cristhian Manuel DuranUniversidad de PamplonaUniversidad de PamplonaOne of the principal inconveniences that analysis and information processing presents is that of the representation of dataset. Normally, one encounters a high number of samples, each one with thousands of variables, and in many cases with irrelevant information and noise. Therefore, in order to represent findings in a clearer way, it is necessary to reduce the amount of variables. In this paper, a novel variable selection technique for multivariable data analysis, inspired on stochastic methods and designed to work with support vector machines (SVM), is described. The approach is demonstrated in a food application involving the detection of adulteration of olive oil (more expensive) with hazelnut oil (cheaper). Fingerprinting by H NMR spectroscopy was used to analyze the different samples. Results show that it is possible to reduce the number of variables without affecting classification results.http://revistas.uis.edu.co/index.php/revistaion/article/view/4594/4814eature selectionH-NMRsimulated annealingsupport vector machineolive oilhazelnut oil
collection	DOAJ
language	English
format	Article
sources	DOAJ
author	Oscar Eduardo Gualdron Claudia Isaza Cristhian Manuel Duran
spellingShingle	Oscar Eduardo Gualdron Claudia Isaza Cristhian Manuel Duran Novel feature selection method based on Stochastic Methods Coupled to Support Vector Machines using H- NMR data (data of olive and hazelnut oils) Revista Ion eature selection H-NMR simulated annealing support vector machine olive oil hazelnut oil
author_facet	Oscar Eduardo Gualdron Claudia Isaza Cristhian Manuel Duran
author_sort	Oscar Eduardo Gualdron
title	Novel feature selection method based on Stochastic Methods Coupled to Support Vector Machines using H- NMR data (data of olive and hazelnut oils)
title_short	Novel feature selection method based on Stochastic Methods Coupled to Support Vector Machines using H- NMR data (data of olive and hazelnut oils)
title_full	Novel feature selection method based on Stochastic Methods Coupled to Support Vector Machines using H- NMR data (data of olive and hazelnut oils)
title_fullStr	Novel feature selection method based on Stochastic Methods Coupled to Support Vector Machines using H- NMR data (data of olive and hazelnut oils)
title_full_unstemmed	Novel feature selection method based on Stochastic Methods Coupled to Support Vector Machines using H- NMR data (data of olive and hazelnut oils)
title_sort	novel feature selection method based on stochastic methods coupled to support vector machines using h- nmr data (data of olive and hazelnut oils)
publisher	Universidad Industrial de Santander
series	Revista Ion
issn	0120-100X 2145-8480
publishDate	2014-12-01
description	One of the principal inconveniences that analysis and information processing presents is that of the representation of dataset. Normally, one encounters a high number of samples, each one with thousands of variables, and in many cases with irrelevant information and noise. Therefore, in order to represent findings in a clearer way, it is necessary to reduce the amount of variables. In this paper, a novel variable selection technique for multivariable data analysis, inspired on stochastic methods and designed to work with support vector machines (SVM), is described. The approach is demonstrated in a food application involving the detection of adulteration of olive oil (more expensive) with hazelnut oil (cheaper). Fingerprinting by H NMR spectroscopy was used to analyze the different samples. Results show that it is possible to reduce the number of variables without affecting classification results.
topic	eature selection H-NMR simulated annealing support vector machine olive oil hazelnut oil
url	http://revistas.uis.edu.co/index.php/revistaion/article/view/4594/4814
work_keys_str_mv	AT oscareduardogualdron novelfeatureselectionmethodbasedonstochasticmethodscoupledtosupportvectormachinesusinghnmrdatadataofoliveandhazelnutoils AT claudiaisaza novelfeatureselectionmethodbasedonstochasticmethodscoupledtosupportvectormachinesusinghnmrdatadataofoliveandhazelnutoils AT cristhianmanuelduran novelfeatureselectionmethodbasedonstochasticmethodscoupledtosupportvectormachinesusinghnmrdatadataofoliveandhazelnutoils
_version_	1725184667529773056

Novel feature selection method based on Stochastic Methods Coupled to Support Vector Machines using H- NMR data (data of olive and hazelnut oils)

Similar Items