Nonlinear partial least squares regressions for spectral quantitative analysis

Abstract As a traditional linear regression model, the partial least squares (PLS) could not handle the nonlinearities in spectral quantitative analysis. This paper focuses on four types of nonlinear partial least square (NPLS) models: the internal NPLS model, the external NPLS model, the nonlinear components extracted NPLS model and the kernel NPLS model. The internal NPLS model adopts the neural network as the nonlinear function to describe the inner relation. For the external NPLS model, the PLS regression is performed on the extended input matrix which contains the nonlinear terms of the independent variables. The nonlinear components extracted NPLS model extracts the nonlinear principal components by selecting the weight vectors with PLS, and then the nonlinear relationship between the nonlinear principal components and the dependent variables is established. For the kernel NPLS model, the original input is transformed into a high-dimensional space by the nonlinear kernel functions and the PLS regression model is built in the new feature space. The 10-fold root–mean–squares error of cross validation is the criterion to decide the optimal parameters of these models. The performance of the different regressions is demonstrated by three real spectral datasets: the meat dataset, the flue gas dataset of gas-fired plant and the flue gas dataset of coal-fired plant. The results suggest that the internal NPLS model with the radial basis function neural network, the external NPLS model with the radial basis function neural network and the kernel NPLS model with polynomial kernel function have a higher predictive ability for spectral quantitative analysis in most cases.

[1]  Application of Near-infrared Spectroscopy in Micro Inorganic Analysis , 2012 .

[2]  Zou Xiaobo,et al.  Variables selection methods in near-infrared spectroscopy. , 2010, Analytica chimica acta.

[3]  Miao Yu,et al.  Rapid Determination of Betulin in Betula platyphylla Outer Bark Using Near-Infrared Spectroscopy , 2013 .

[4]  S. D. Jong.  SIMPLS: an alternative approach to partial least squares regression , 1993 .

[5]  Ildiko E. Frank,et al.  Modern nonlinear regression methods , 1995 .

[6]  A. J. Morris,et al.  Non-linear projection to latent structures revisited (the neural network PLS algorithm) , 1999 .

[7]  Wei-Qi Lin,et al.  Artificial neural network-based transformation for nonlinear partial least-square regression with application to QSAR studies. , 2007, Talanta.

[8]  M. Rantalainen,et al.  Kernel‐based orthogonal projections to latent structures (K‐OPLS) , 2007 .

[9]  Shi-Miao Tan,et al.  Quantitative analysis of tea using ytterbium‐based internal standard near‐infrared spectroscopy coupled with boosting least‐squares support vector regression , 2013 .

[10]  S. Wold,et al.  Nonlinear PLS modeling , 1989 .

[11]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[12]  Lian Xie,et al.  On-line dissolution determination of Baicalin in solid dispersion based on near infrared spectroscopy and circulation dissolution system , 2011 .

[13]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[14]  M. Hubert,et al.  Robust methods for partial least squares regression , 2003 .

[15]  Erkki Oja,et al.  Rival penalized competitive learning for clustering analysis, RBF net, and curve detection , 1993, IEEE Trans. Neural Networks.

[16]  T. McAvoy,et al.  Nonlinear principal component analysis—Based on principal curves and neural networks , 1996 .

[17]  Desire L. Massart,et al.  Local modelling with radial basis function networks , 2000 .

[18]  Chonghun Han,et al.  Robust Nonlinear PLS Based on Neural Networks and Application to Composition Estimator for High-Purity Distillation Columns , 2000 .

[19]  Beata Walczak,et al.  Non-linear modelling of chemical data by combinations of linear and neural net methods , 1993 .

[20]  B. Kowalski,et al.  Partial least-squares regression: a tutorial , 1986 .

[21]  H. H. Thodberg,et al.  Optimal minimal neural interpretation of spectra , 1992 .

[22]  Roman Rosipal,et al.  Kernel Partial Least Squares Regression in Reproducing Kernel Hilbert Space , 2002, J. Mach. Learn. Res..

[23]  S. Qin.  Recursive PLS algorithms for adaptive data modeling , 1998 .

[24]  Lutgarde M. C. Buydens,et al.  Aspects of multi-layer feed-forward neural networks influencing the quality of the fit of univariate non-linear relationships , 1995 .

[25]  Miguel de la Guardia,et al.  Modified locally weighted--partial least squares regression improving clinical predictions from infrared spectra of human serum samples. , 2013, Talanta.

[26]  R. M. Hoffman,et al.  Kinetic modeling of dissolution and crystallization of slurries with attenuated total reflectance UV-visible absorbance and near-infrared reflectance measurements. , 2013, Analytical chemistry.

[27]  S. Wold,et al.  Some recent developments in PLS modeling , 2001 .

[28]  T. Næs,et al.  Locally weighted regression and scatter correction for near-infrared reflectance data , 1990 .

[29]  M. Kramer.  Nonlinear principal component analysis using autoassociative neural networks , 1991 .

[30]  L. Buydens,et al.  Multivariate calibration with least-squares support vector machines. , 2004, Analytical chemistry.

[31]  Masafumi Hagiwara,et al.  Smart chemical taste sensor for determination and prediction of taste qualities based on a two-phase optimized radial basis function network. , 2005, Analytical chemistry.

[32]  John Moody,et al.  Fast Learning in Networks of Locally-Tuned Processing Units , 1989, Neural Computation.

[33]  P. Hopke,et al.  Classification of Single Particles Analyzed by ATOFMS Using an Artificial Neural Network, ART-2A , 1999 .

[34]  M. Mørup,et al.  Non-linear calibration models for near infrared spectroscopy. , 2014, Analytica chimica acta.

[35]  Andrew Rawson,et al.  The prediction of soil chemical and physical properties from mid-infrared spectroscopy and combined partial least-squares regression and neural networks (PLS-NN) analysis , 2009 .

[36]  G. Lim,et al.  A nonlinear partial least squares algorithm using quadratic fuzzy inference system , 2009 .

[37]  Adam Buciński,et al.  Artificial neural network analysis for evaluation of peptide MS/MS spectra in proteomics. , 2004, Analytical chemistry.

[38]  Gang Li,et al.  Composition Analysis of Scattering Liquids Based on Spatially Offset Visible-Near-Infrared Spectroscopy , 2012, Applied spectroscopy.

[39]  J. Macgregor,et al.  Product optimization and control in the latent variable space of nonlinear PLS models , 2004 .

[40]  S. Wold,et al.  The kernel algorithm for PLS , 1993 .

[41]  Yizeng Liang,et al.  Exploring nonlinear relationships in chemical data using kernel-based methods , 2011 .

[42]  Manfred Morari,et al.  PLS/neural networks , 1992 .

[43]  Erik Johansson,et al.  Unlocking interpretation in near infrared multivariate calibrations by orthogonal partial least squares. , 2009, Analytical chemistry.

[44]  Priyanka Ojha,et al.  Partial least squares and artificial neural networks modeling for predicting chlorophenol removal from aqueous solution , 2009 .

[45]  Dusan Stulik,et al.  Simple encoding of infrared spectra for pattern recognition Part 2. Neural network approach using back-propagation and associative Hopfield memory , 1995 .

[46]  S. Peng,et al.  Partial least squares-slice transform hybrid model for nonlinear calibration , 2014 .

[47]  B. Álvarez-Sánchez,et al.  Near‐infrared spectroscopy and partial least squares‐class modeling (PLS‐CM) for metabolomics fingerprinting discrimination of intervention breakfasts ingested by obese individuals , 2013 .

[48]  Z. Boger.  Selection of quasi-optimal inputs in chemometrics modeling by artificial neural network analysis , 2003 .

[49]  S. Wold.  Nonlinear partial least squares modelling II. Spline inner relation , 1992 .

[50]  H. Goicoechea,et al.  Monitoring substrate and products in a bioprocess with FTIR spectroscopy coupled to artificial neural networks enhanced with a genetic-algorithm-based method for wavelength selection. , 2006, Talanta.

[51]  I. E. Frank.  A nonlinear PLS model , 1990 .

[52]  E. Martin,et al.  Non-linear projection to latent structures revisited: the quadratic PLS algorithm , 1999 .

[53]  Bart Nicolai,et al.  Kernel PLS regression on wavelet transformed NIR spectra for prediction of sugar content of apple , 2007 .

[54]  P Dardenne,et al.  Comparison of various chemometric approaches for large near infrared spectroscopic data of feed and feed products. , 2011, Analytica chimica acta.

[55]  Thomas J. McAvoy,et al.  Nonlinear PLS Modeling Using Neural Networks , 1992 .

[56]  Zhizhong Mao,et al.  Kernel partial robust M-regression as a flexible robust nonlinear modeling technique , 2010 .

[57]  Yukihiro Ozaki,et al.  Investigations of bagged kernel partial least squares (KPLS) and boosting KPLS with applications to near‐infrared (NIR) spectra , 2006 .

[58]  Edward C. Malthouse,et al.  Nonlinear partial least squares , 1997 .

[59]  D. Massart,et al.  The Radial Basis Functions — Partial Least Squares approach as a flexible non-linear regression technique , 1996 .