Journal of Chemometrics

Cover image for Vol. 29 Issue 5

Early View (Online Version of Record published before inclusion in an issue)

Edited By: Prof. Paul J Gemperline

Impact Factor: 1.803

ISI Journal Citation Reports © Ranking: 2013: 15/57 (Instruments & Instrumentation); 15/119 (Statistics & Probability); 20/95 (Mathematics Interdisciplinary Applications); 23/59 (Automation & Control Systems); 39/121 (Computer Science Artificial Intelligence); 40/76 (Chemistry Analytical)

Online ISSN: 1099-128X


  1. 1 - 9
  1. Research articles

    1. In silico evaluation of logD7.4 and comparison with other prediction methods

      Jian-Bing Wang, Dong-Sheng Cao, Min-Feng Zhu, Yong-Huan Yun, Nan Xiao and Yi-Zeng Liang

      Article first published online: 25 MAY 2015 | DOI: 10.1002/cem.2718

      As an important parameter in medicinal chemistry, the evaluation of logD7.4 is of high importance. Here, we developed a quantitative structure–property relationship model to reliably predict the logD7.4 with a big and diverse data set. A series of evaluation steps such as cross validation, Y-randomization test, applicability domain, and the external test demonstrate the robustness and reliability of our model. When compared with several calculation methods from ChemAxon and Discovery Studio, the support vector machine model also shows superiority over them.

    2. N-way modeling for wavelet filter determination in multivariate image analysis

      José Manuel Prats-Montalbán, Marina Cocchi and Alberto Ferrer

      Article first published online: 12 MAY 2015 | DOI: 10.1002/cem.2717

      Wavelets are one of the state-of-the-art tools in texture image analysis. However, image analysis is problem-dependent, and different applications might require different wavelets in order to gather the main sources of variation in the images with respect to the specific task to be performed.

      This paper provides a methodology based on N-way modeling for properly selecting a range of possible wavelet choices (in terms of families, filters, and decomposition levels), for each image and problem at hand.

  2. Special issue articles

    1. Fault diagnosis using kNN reconstruction on MRI variables

      Guozhu Wang, Jianchang Liu and Yuan Li

      Article first published online: 5 MAY 2015 | DOI: 10.1002/cem.2719

      A novel fault diagnosis method is derived using k-nearest neighbor reconstruction on maximize reduce index (MRI) sensors; it is aimed at identifying all fault variables precisely. This method can identify the faulty variables effectively through reconstructing MRI variables one by one.

    2. Sparse canonical correlation analysis applied to -omics studies for integrative analysis and biomarker discovery

      Dong-Sheng Cao, Shao Liu, Wen-Bin Zeng and Yi-Zeng Liang

      Article first published online: 19 APR 2015 | DOI: 10.1002/cem.2716

      With the rapid development of new -omics measurement methods, there is an increasing interest in studying the correlation structure between two or more data sets. Here, we explored the potential of sparse CCA to find the correlative components in two sparse views. SCCA aims at finding sparse projection directions to well extract the correlation between two data sets. We applied this method to one simulation data and one real -omics data to illustrate the performance of SCCA.

  3. Research articles

    1. Dynamic mixture probabilistic PCA classifier modeling and application for fault classification

      Jinlin Zhu, Zhiqiang Ge and Zhihuan Song

      Article first published online: 31 MAR 2015 | DOI: 10.1002/cem.2714

      A dynamic classifier based on the mixture probabilistic principal component analyzer (MPPCA) is proposed for fault classification. By introducing a state indicator, conventional MPPCA is designed as a classifier. Then, the static MPPCA classifier is temporally extended to the dynamic form within the hidden Markov model framework. The dynamic MPPCA classifier is obtained by Expectation-Maximization algorithm. In simulation, case studies of the continuous stirred tank heater process and the Tennessee Eastman process are carried out.

    2. A Bayesian sparse reconstruction method for fault detection and isolation

      Jiusun Zeng, Biao Huang and Lei Xie

      Article first published online: 17 MAR 2015 | DOI: 10.1002/cem.2712

      A Bayesian sparse reconstructions method is proposed for fault detection and isolation. A simple singal-plus-noise model is used for both normal and test data. An indicator matrix is introduced to show whether the test data is faulty. Laplacian prior is imposed on the indicator matrix, which forces the matrix to be sparse. Gibbs sampling is used to obtain the estimation of parameters.

    3. Chemometrics-enhanced laser-induced thermal emission detection of PETN and other explosives on various substrates

      Amanda M. Figueroa-Navedo, Nataly J. Galán-Freyle, Leonardo C. Pacheco-Londoño and Samuel P. Hernández-Rivera

      Article first published online: 17 MAR 2015 | DOI: 10.1002/cem.2704

      Laser-induced thermal emission remote detection of substrates with explosives residues was highly enhanced by chemometrics analyses. Emissions with/without pentaerythritol tetranitrate were used to generate models for discrimination. Principal component analysis, soft independent modeling by class analogy (SIMCA), partial least squares-discriminant analysis (PLS-DA), support vector machines (SVM), and neural networks (NN) analyses were performed. Prediction accuracies were SVM, PLS-DA, NN, and SIMCA: 95%, 94%, 88% and 45%, respectively. High sensitivities and specificities were achieved for five of the seven substrates investigated.

    4. Evaluation of smoothing techniques in the run to run optimization of fed-batch processes with u-PLS

      José Camacho, David Lauri, Barry Lennox, Manolo Escabias and Mariano Valderrama

      Article first published online: 11 MAR 2015 | DOI: 10.1002/cem.2711

      Run to run (R2R) optimization based on unfolded Partial Least Squares (u-PLS) is a promising approach for improving the performance of batch and fed-batch processes. The optimization performs better when PLS is combined with a smoothing technique. In this paper, the suitability of different smoothing techniques in combination with PLS is studied for both end-of-batch quality prediction and R2R optimization, including filtering, the introduction of smoothing constraints in the PLS calibration (Penalized PLS), and functional analysis (Functional PLS).

  4. The chemometrics column

    1. Populations and samples

      Richard G. Brereton

      Article first published online: 5 MAR 2015 | DOI: 10.1002/cem.2695


  1. 1 - 9