Support vector machine regression (SVR/LS-SVM)—an alternative to neural networks (ANN) for analytical chemistry? Comparison of nonlinear methods on near infrared (NIR) spectroscopy data

The Analyst ◽  
2011 ◽  
Vol 136 (8) ◽  
pp. 1703 ◽  
Author(s):  
Roman M. Balabin ◽  
Ekaterina I. Lomakina
Keyword(s):  
Analytical Chemistry ◽  
Neural Networks ◽  
Support Vector Machine ◽  
Near Infrared ◽  
Nir Spectroscopy ◽  
Support Vector ◽  
Spectroscopy Data ◽  
Support Vector Machine Regression ◽  
Nonlinear Methods

scholarly journals A New Plant Indicator (Artemisia lavandulaefolia DC.) of Mercury in Soil Developed by Fourier-Transform Near-Infrared Spectroscopy Coupled with Least Squares Support Vector Machine

2019 ◽  
Vol 2019 ◽  
pp. 1-6
Author(s):  
Lu Xu ◽  
Qiong Shi ◽  
Bang-Cheng Tang ◽  
Shunping Xie
Keyword(s):  
Support Vector Machine ◽  
Fourier Transform ◽  
Least Squares ◽  
Near Infrared ◽  
Mercury Level ◽  
Reference Method ◽  
Nir Spectroscopy ◽  
Classification Model ◽  
Support Vector ◽  
Reference Analysis

A rapid indicator of mercury in soil using a plant (Artemisia lavandulaefolia DC., ALDC) commonly distributed in mercury mining area was established by fusion of Fourier-transform near-infrared (FT-NIR) spectroscopy coupled with least squares support vector machine (LS-SVM). The representative samples of ALDC (stem and leaf) were gathered from the surrounding and distant areas of the mercury mines. As a reference method, the total mercury contents in soil and ALDC samples were determined by a direct mercury analyzer incorporating high-temperature decomposition, catalytic adsorption for impurity removal, amalgamation capture, and atomic absorption spectrometry (AAS). Based on the FT-NIR data of ALDC samples, LS-SVM models were established to distinguish mercury-contaminated and ordinary soil. The results of reference analysis showed that the mercury level of the areas surrounding mercury mines (0–3 kilometers, 7.52–88.59 mg/kg) was significantly higher than that of the areas distant from mercury mines (>5 kilometers, 0–0.75 mg/kg). The LS-SVM classification model of ALDC samples was established based on the original spectra, smoothed spectra, second-derivative (D2) spectra, and standard normal transformation (SNV) spectra, respectively. The prediction accuracy of D2-LS-SVM was the highest (0.950). FT-NIR combined with LS-SVM modeling can quickly and accurately identify the contaminated ALDC. Compared with traditional methods which rely on naked eye observation of plants, this method is objective and more sensitive and applicable.

scholarly journals PERFORMANCE OF THE SUPPORT VECTOR MACHINE AND ARTIFICIAL NEURAL NETWORK CLASSIFIERS FOR ROADS IDENTIFICATION

2021 ◽  
Vol XLIII-B3-2021 ◽  
pp. 9-14
Author(s):  
A. C. Andrade ◽  
M. J. Alixandrini Jr. ◽  
F. P. S. Carvalho ◽  
V. O. Fernandes
Keyword(s):  
Neural Networks ◽  
Support Vector Machine ◽  
Laser Scanning ◽  
Near Infrared ◽  
Input Image ◽  
Support Vector ◽  
Surface Model ◽  
Infrared Band ◽  
Artificial Neural ◽  
Non Parametric

Abstract. The objective of this project was to compare two non-parametric classification methods (“Support Vector Machine” – SVM and “Artificial Neural Networks” – ANN) of road regions in high spatial resolution images and associated with data from Airborne Laser Scanning. The study aims to verify what kind of influence the layers of attributes have on the performance from respective classifiers: SVM and RNA. Our method based on tests of this classifiers on 4 bands of airborne images and normalization of the digital surface model (DSM) for showing only information on objects height in relation to ground and not of these in relation to the ground and relief, generating band 5. The samples were used to train chosen non-parametric classifiers (training sets for each different input image/landscape). All classifications had the same set of training samples and the same classification parameters. The optimal parameters for classifications were obtained through the existing library in the Weka mining package: LibSVM and LibMultiLayerPerceptron. Our results demonstrated the existence of a direct relationship between the elevation band of the targets in relation to the terrain (band 05) with the improvement of their performance and lower degree of between bands correlation can also be considered a factor that has a positive influence. As for Neural Networks, the experiment results demonstrate that the presence of the near infrared band (band 04) was decisive for the performance improving of certain combinations in relation to others.

scholarly journals Discriminating the Geographical Origins of Chinese White Lotus Seeds by Near-Infrared Spectroscopy and Chemometrics

2015 ◽  
Vol 2015 ◽  
pp. 1-8
Author(s):  
Lu Xu ◽  
Chen-Bo Cai ◽  
Yuan-Bin She ◽  
Li-Juan Chen
Keyword(s):  
Support Vector Machine ◽  
Infrared Spectroscopy ◽  
Least Squares ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Nir Spectroscopy ◽  
Support Vector ◽  
Discrimination Analysis ◽  
Protected Designation Of Origin ◽  
Chinese White

The traceability of a Chinese white lotus seed (WLS) with Protected Designation of Origin (PDO) was investigated using near-infrared (NIR) spectroscopy and chemometrics. Three chemometrics methods, discrimination analysis (DA), class modeling, and a newly proposed strategy, the fusion of DA and class modeling, were investigated to compare their capacity to trace the geographical origins of WLS. Least squares support vector machine (LS-SVM) was developed to distinguish the PDO WLS from non-PDO WLS of four main producing areas. A class modeling technique, one-class partial least squares (OCPLS), was developed only using the data of PDO WLS. By the fusion of LS-SVM and OCPLS, the best prediction sensitivity and specificity were 0.900 and 0.973, respectively. The results indicate that fusion of DA and class modeling can enhance the specificity for detection of non-PDO products. The conclusion is that DA and class modeling should be combined for tracing food geographical origins.

Rapid Detection of Melamine in Milk Powder by near Infrared Spectroscopy

2009 ◽  
Vol 17 (2) ◽  
pp. 59-67 ◽  
Author(s):  
Chenghui Lu ◽  
Bingren Xiang ◽  
Gang Hao ◽  
Jianping Xu ◽  
Zhengwu Wang ◽  
...  
Keyword(s):  
Support Vector Machine ◽  
Classification Accuracy ◽  
Rapid Detection ◽  
Near Infrared ◽  
Milk Powder ◽  
Nir Spectroscopy ◽  
Accurate Method ◽  
Partial Least Square ◽  
Least Square ◽  
Support Vector

This paper establishes a novel and rapid method for detecting pure melamine in milk powder using near infrared (NIR) spectroscopy based on least squares-support vector machine (LS-SVM). Partial least square discriminant analysis (PLS-DA) was used for the extraction of principal components (PCs). The scores of the first two PCs have been applied as inputs to LS-SVM. Compared to PLS-DA, the performance of LS-SVM was better, with higher classification accuracy, both 100% for the training and testing set. The detection limit was lower than 1 ppm. Based on the results, it was concluded that NIR spectroscopy combined with LS-SVM could be used as a rapid and accurate method for detecting pure melamine in milk powder.

Application of Support Vector Machine in Identifying inside Moldy Chestnut Using Near Infrared Spectroscopy

2014 ◽  
Vol 615 ◽  
pp. 169-172
Author(s):  
Jie Liu ◽  
Xiao Yu Li ◽  
Wei Wang ◽  
Jun Zhang
Keyword(s):  
Support Vector Machine ◽  
Near Infrared ◽  
Scatter Correction ◽  
Nir Spectroscopy ◽  
Support Vector ◽  
Training Set ◽  
Near Infrared Spectra ◽  
Standard Normal ◽  
Validation Set ◽  
Normal Transformation

NIR spectroscopy has been applied in detecting inside quality of chestnut successfully. In this work, Support Vector Machine Discriminant Analysis was utilized to identify the qualified chestnuts, the serious moldy chestnuts and the slight moldy chestnuts using their Near infrared spectra region from 833 nm to 2500 nm. 109 chestnut samples were involved and four different preprocessing methods were compared. The results showed that for all the models, the average correct rates of training set and validation set were higher than 90%. The performance of model based on raw spectra was not as good as other models, which indicated the necessity of preprocessing. The models based on the spectra preprocessed by first derivative and multiplicative scatter correction had the same performances, with 97% and 85% as the correct rate of training set and validation set. The models based on the spectra preprocessed by Standard normal transformation has 100% correct rate of training set while 88% of validation set. The second derivative model had the best result with 100% and 90% as the correct rate of training set and validation set. These results demonstrated that the NIR spectroscopy had capability to detect interior mildew of intact chestnut nondestructively.

Fourier Transform Near-Infrared Spectroscopy and Chemometrics To Predict Zygosacchromyces rouxii in Apple and Kiwi Fruit Juices

2018 ◽  
Vol 81 (8) ◽  
pp. 1379-1385 ◽  
Author(s):  
CHEN NIU ◽  
HONG GUO ◽  
JIANPING WEI ◽  
MARINA SAJID ◽  
YAHONG YUAN ◽  
...  
Keyword(s):  
Support Vector Machine ◽  
Infrared Spectroscopy ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Fruit Juices ◽  
Support Vector ◽  
Support Vector Machine Regression ◽  
Limit Of Quantification ◽  
Kiwi Fruit ◽  
Single Variety

ABSTRACT This study investigated the capability of near-infrared spectroscopy (NIRS) to predict the concentration of Zygosaccharomyces rouxii in apple and kiwi fruit juices. The yeast was inoculated in fresh kiwi fruit juice (n = 68), reconstituted kiwi juice (n = 85), and reconstituted apple juice (n = 64), followed by NIR spectra collection and plate counting. A principal component analysis indicated direct orthogonal signal correction preprocessing was suitable to separate spectral samples. Parameter optimization algorithms increased the performance of support vector machine regression models developed in a single variety juice system and a multiple variety juice system. Single variety juice models achieved accurate prediction of Z. rouxii concentrations, with the limit of quantification at 3 to 15 CFU/mL (R2 = 0.997 to 0.999), and the method was also feasible for Hanseniaspora uvarum and Candida tropicalis. The best multiple variety juice model obtained had a limit of quantification of 237 CFU/mL (R2 = 0.961) for Z. rouxii. A Bland-Altman analysis indicated good agreement between the support vector machine regression model and the plate counting method. It suggests that NIRS can be a high-throughput method for prediction of Z. rouxii counts in kiwi fruit and apple juices.

Sign in / Sign up

Export Citation Format

Share Document