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
Top Cited Papers
- 25 February 2011
- journal article
- research article
- Published by Royal Society of Chemistry (RSC) in The Analyst
- Vol. 136 (8) , 1703-1712
Abstract
In this study, we make a general comparison of the accuracy and robustness of five multivariate calibration models: partial least squares (PLS) regression or projection to latent structures, polynomial partial least squares (Poly-PLS) regression, artificial neural networks (ANNs), and two novel techniques based on support vector machines (SVMs) for multivariate data analysis: support vector regression (SVR) and least-squares support vector machines (LS-SVMs). The comparison is based on fourteen (14) different datasets: seven sets of gasoline data (density, benzene content, and fractional composition/boiling points), two sets of ethanol gasoline fuel data (density and ethanol content), one set of diesel fuel data (total sulfur content), three sets of petroleum (crude oil) macromolecules data (weight percentages of asphaltenes, resins, and paraffins), and one set of petroleum resins data (resins content). Vibrational (near-infrared, NIR) spectroscopic data are used to predict the properties and quality coefficients of gasoline, biofuel/biodiesel, diesel fuel, and other samples of interest. The four systems presented here range greatly in composition, properties, strength of intermolecular interactions (e.g., van der Waals forces, H-bonds), colloid structure, and phase behavior. Due to the high diversity of chemical systems studied, general conclusions about SVM regression methods can be made. We try to answer the following question: to what extent can SVM-based techniques replace ANN-based approaches in real-world (industrial/scientific) applications? The results show that both SVR and LS-SVM methods are comparable to ANNs in accuracy. Due to the much higher robustness of the former, the SVM-based approaches are recommended for practical (industrial) application. This has been shown to be especially true for complicated, highly nonlinear objects.Keywords
This publication has 62 references indexed in Scilit:
- Adsorption of Petroleum Asphaltenes onto Reservoir Rock Sands Studied by Near-Infrared (NIR) SpectroscopyEnergy & Fuels, 2009
- Wavelet neural network (WNN) approach for calibration model building based on gasoline near infrared (NIR) spectraChemometrics and Intelligent Laboratory Systems, 2008
- Petroleum resins adsorption onto quartz sand: Near infrared (NIR) spectroscopy studyJournal of Colloid and Interface Science, 2008
- Capabilities of near Infrared Spectroscopy for the Determination of Petroleum Macromolecule Content in Aromatic SolutionsJournal of Near Infrared Spectroscopy, 2007
- Comparison of linear and nonlinear calibration models based on near infrared (NIR) spectroscopy data for gasoline properties predictionChemometrics and Intelligent Laboratory Systems, 2007
- Gasoline quality prediction using gas chromatography and FTIR spectroscopy: An artificial intelligence approachFuel, 2006
- Comparing support vector machines to PLS for spectral regression applicationsChemometrics and Intelligent Laboratory Systems, 2004
- Multivariate Calibration with Least-Squares Support Vector MachinesAnalytical Chemistry, 2004
- Application of LS-SVM to non-linear phenomena in NIR spectroscopy: development of a robust and portable sensor for acidity prediction in grapesChemometrics and Intelligent Laboratory Systems, 2004
- Least-Squares Support Vector Machines for Chemometrics: An Introduction and EvaluationJournal of Near Infrared Spectroscopy, 2004