Journal of Chemometrics

Cover image for Vol. 27 Issue 12

December 2013

Volume 27, Issue 12

Pages i–v, 431–487

  1. Issue Information

    1. Top of page
    2. Issue Information
    3. Cover Image
    4. Special Feature
    5. Editorial
    6. Review
    7. Research Articles
    1. Issue Information (pages i–iii)

      Article first published online: 8 DEC 2013 | DOI: 10.1002/cem.2477

  2. Cover Image

    1. Top of page
    2. Issue Information
    3. Cover Image
    4. Special Feature
    5. Editorial
    6. Review
    7. Research Articles
    1. Cover Image (page iv)

      Article first published online: 22 OCT 2013 | DOI: 10.1002/cem.2564

      Thumbnail image of graphical abstract
  3. Special Feature

    1. Top of page
    2. Issue Information
    3. Cover Image
    4. Special Feature
    5. Editorial
    6. Review
    7. Research Articles
    1. Probabilistic model-based discriminant analysis and clustering methods in chemometrics (page v)

      Charles Bouveyron

      Article first published online: 6 NOV 2013 | DOI: 10.1002/cem.2563

      Thumbnail image of graphical abstract

      In Chemometrics, the supervised and unsupervised classification of high-dimensional data has become a recurrent problem. Model-based techniques for discriminant analysis and clustering are popular tools which are renowned for their probabilistic foundations and their flexibility. However, classical model-based techniques show a disappointing behavior in high-dimensional spaces which up to now have been limited in their use within Chemometrics. The recent developments in model-based classification overcame these drawbacks and enabled the efficient classification of high-dimensional data. This work presents a comprehensive review of these recent approaches, including regularization-based techniques, parsimonious modeling, subspace classification methods and classification methods based on variable selection.

  4. Editorial

    1. Top of page
    2. Issue Information
    3. Cover Image
    4. Special Feature
    5. Editorial
    6. Review
    7. Research Articles
  5. Review

    1. Top of page
    2. Issue Information
    3. Cover Image
    4. Special Feature
    5. Editorial
    6. Review
    7. Research Articles
    1. You have free access to this content
      Probabilistic model-based discriminant analysis and clustering methods in chemometrics (pages 433–446)

      Charles Bouveyron

      Article first published online: 16 OCT 2013 | DOI: 10.1002/cem.2560

      In chemometrics, the supervised and unsupervised classifications of high-dimensional data have become a recurrent problem. Model-based techniques for discriminant analysis and clustering are popular tools that are renowned for their probabilistic foundations and their flexibility. The recent developments in model-based classification overcame these drawbacks and enabled the efficient classification of high-dimensional data. This work presents a comprehensive review of these recent approaches, including regularization-based techniques, parsimonious modelling, subspace classification methods and classification methods based on variable selection.

  6. Research Articles

    1. Top of page
    2. Issue Information
    3. Cover Image
    4. Special Feature
    5. Editorial
    6. Review
    7. Research Articles
    1. A novel fusion approach based on induced ordered weighted averaging operators for chemometric data analysis (pages 447–456)

      Haikel AlHichri, Yakoub Bazi, Naif Alajlan, Farid Melgani, Salim Malek and Ronald R. Yager

      Article first published online: 6 OCT 2013 | DOI: 10.1002/cem.2557

      This paper proposes a novel approach for the estimation of spectroscopic data by combining the predictions of an ensemble of estimators using the induced ordered weighted averaging (IOWA) fusion operators. For ensemble generation, we use Gaussian process regression and extreme learning machine estimators associated with different kernels. For the IOWA operator, we use an estimation of the residual as an order-inducing value, and the concept of prioritized aggregation to generate the weights. Experimental results are presented and discussed.

    2. Supervised principal components: a new method for multivariate spectral analysis (pages 457–465)

      Jun Bin, Fang-Fang Ai, Nian Liu, Zhi-Min Zhang, Yi-Zeng Liang, Ru-Xin Shu and Kai Yang

      Article first published online: 9 OCT 2013 | DOI: 10.1002/cem.2558

      ▸We applied supervised principal components method to multivariate spectral analysis. ▸A score statistic and a log-likelihood ratio statistic were employed to choose those correct variables. ▸Supervised principal components can reduce the risk of overfitting and the effect of collinearity in modeling with a preferable ability of estimation because of its semi-supervised strategy.

    3. Comparison of interpolation polynomials with divided differences, interpolation polynomials with finite differences, and quadratic functions obtained by the least squares method in modeling of chromatographic responses (pages 466–474)

      Tijana Rakić, Zorica Stanimirović, Aleksandar Đenić, Miroslav Marić, Biljana Jančić-Stojanović and Mirjana Medenica

      Article first published online: 7 OCT 2013 | DOI: 10.1002/cem.2559

      Modeling of chromatographic responses based on interpolation polynomial with divided differences, finite differences, and quadratic function is compared. Novel techniques are incorporated in Design of Experiments methodology for systematical development and optimization of liquid chromatographic method. Interpolation polynomial with divided differences succeeded to locate optimum and high agreement between theoretical and experimental chromatograms is obtained. Interpolation polynomial with finite differences and quadratic function failed to locate the optimum.

    4. Wood identification using pressure DSC data (pages 475–487)

      Javier Tarrío-Saavedra, Mario Francisco-Fernández, Salvador Naya, Jorge López-Beceiro, Carlos Gracia-Fernández and Ramón Artiaga

      Article first published online: 24 OCT 2013 | DOI: 10.1002/cem.2561

      Pressure differential scanning calorimetry (PDSC) represents a new way to obtain thermo-oxidative information from wood species. A supervised classification problem of wood species, applying functional data analysis (FDA) and multivariate techniques to PDSC data, is performed. The FDA techniques are competitive with respect to partial linear squares (PLS) and principal component analysis (PCA) approaches. New PDSC fit model parameters and six estimators of the fractal dimension were proposed as an alternative to PLS/PCA features extraction. The supervised classification by PDSC curves has been shown to be feasible, fast, and robust.

SEARCH

SEARCH BY CITATION