Journal of Chemometrics最新文献

{"title":"Complementary Chemometrics Analysis for Processing Rivers Monitoring Data","authors":"María L. Rolandi,&nbsp;Nora M. Urquiza,&nbsp;Alejandro G. García-Reiriz","doi":"10.1002/cem.70125","DOIUrl":"10.1002/cem.70125","url":null,"abstract":"<div>\u0000 \u0000 <p>This research underscores the value of integrating multiple chemometric techniques to optimize the analysis and interpretation of complex environmental datasets, specifically in the context of river pollution assessment. Applying a complementary suite of methods—principal component analysis (PCA), bilinear multivariate curve resolution with alternating least squares (MCR-ALS) and trilinear MCR-ALS—this study effectively disentangles natural geochemical patterns and diffuse inputs, revealing their spatiotemporal dynamics within Tucumán's river systems. Bimonthly water sampling across 10 sites over 2 years generated a rich multivariate dataset, which was systematically processed and analysed. The combination of methodologies allowed for the identification of different geochemical patterns, including two natural origins—one characterized by bicarbonate and alkaline earth ions linked to lithological weathering and another associated with pH and acid-base equilibria modulated by hydrology—as well as two nutrient enrichment patterns, characterized by nutrients (NO₂⁻, NO₃⁻ and PO₄³⁻) and chlorides, with accumulation downstream in areas influenced by agriculture and urbanization. Riparian vegetation played a mitigating role, with sites in better conservation demonstrating a more effective attenuation of these impacts. The combined use of these chemometric tools underscores their complementary strengths, enhancing understanding of diffuse pollution dynamics and supporting informed environmental management decisions.</p>\u0000 </div>","PeriodicalId":15274,"journal":{"name":"Journal of Chemometrics","volume":"40 4","pages":""},"PeriodicalIF":2.1,"publicationDate":"2026-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147683557","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Extended Multiplicative Signal Correction of Noisy Raman Spectra","authors":"Evgeniy G. Evtushenko,&nbsp;Ilya N. Kurochkin","doi":"10.1002/cem.70123","DOIUrl":"10.1002/cem.70123","url":null,"abstract":"<div>\u0000 \u0000 <p>Extended multiplicative signal correction (EMSC) is a common toolkit for preprocessing of vibrational spectra in chemometrics. The core of the EMSC is the ordinary least squares procedure, which is known to be affected by regression dilution bias in the presence of random noise in independent variables, i.e., reference spectra. For the simplest EMSC instance containing a single noisy reference spectrum with several baseline components, we introduce the proper statistical model for the EMSC procedure, which shows that the array of processed spectra is divided into two sets with different bias properties. An experimentally assessable parameter <i>α</i>, which defines the magnitude of bias for both sets, was suggested. The validity of several existing estimators together with two newly introduced ones was theoretically evaluated for bias correction. Next, selected estimators were tested using a specially designed experimental data set of sucrose Raman spectra. We believe that our study will serve as a proper introduction to the broad scientific field of regression dilution bias management for EMSC procedures.</p>\u0000 </div>","PeriodicalId":15274,"journal":{"name":"Journal of Chemometrics","volume":"40 4","pages":""},"PeriodicalIF":2.1,"publicationDate":"2026-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147683348","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Support Vector Machine Ensemble for Crude Oil Classification","authors":"Maria Carolina da V. A. Barboza,&nbsp;Márcia Helena Cassago Nascimento,&nbsp;Gabriely Silveira Folli,&nbsp;Junio R. Botelho,&nbsp;Rafael de Queiroz Ferreira,&nbsp;Wanderson Romão,&nbsp;Paulo Roberto Filgueiras","doi":"10.1002/cem.70119","DOIUrl":"https://doi.org/10.1002/cem.70119","url":null,"abstract":"<p>The study focuses on proposing a new machine learning approach to classify crude oil samples based on their physicochemical properties, such as sulfur concentration (S), total acidity number (TAN), and American Petroleum Institute (API) gravity. The goal is to overcome the limitations of traditional analysis methods, which are time-consuming and consume large volumes of samples and solvents, using spectroscopic techniques and machine learning models such as support vector machine ensemble (SVM ensemble). The number of 196 crude oil samples with different sulfur content, different TAN, and API gravity were considered. The SVM ensemble is a powerful approach to improve classification performance because it can reduce the variability of individual models, improve robustness against overfitting, and generalize better than a single model. The parameters of sensitivity, specificity, error rate, Matthews correlation coefficient, and accuracy were considered to compare the SVM ensemble with partial least squares-discriminant analysis (PLS-DA) and SVM. The results demonstrated that near infrared spectroscopy (NIR), combined with multivariate classification models, is an efficient and reliable method for simultaneously classifying sulfur content, TAN, and API gravity in crude oils.</p>","PeriodicalId":15274,"journal":{"name":"Journal of Chemometrics","volume":"40 4","pages":""},"PeriodicalIF":2.1,"publicationDate":"2026-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/epdf/10.1002/cem.70119","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147569085","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Why Is the Peak Group Analysis So Effective for IR Spectra Analysis?","authors":"Klaus Neymeyr,&nbsp;Christoph Kubis,&nbsp;Lukas Prestin,&nbsp;Robert Franke,&nbsp;Mathias Sawall","doi":"10.1002/cem.70121","DOIUrl":"https://doi.org/10.1002/cem.70121","url":null,"abstract":"<p>Peak group analysis (PGA), an MCR algorithm, has proven to be very successful in extracting pure component spectra and concentration profiles from IR and Raman spectral data when investigating carbonylation reactions with transition metal catalysts. In this field of study, mixture spectra typically exhibit high spectral selectivity, meaning certain peaks belong exclusively to a single chemical species. Under this condition, PGA can extract the associated pure component spectrum from the mixture spectra, and there is no factor ambiguity in these profiles. Here, we present a short mathematical proof of this remarkable PGA property.</p>","PeriodicalId":15274,"journal":{"name":"Journal of Chemometrics","volume":"40 4","pages":""},"PeriodicalIF":2.1,"publicationDate":"2026-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/epdf/10.1002/cem.70121","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147568789","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correlation-Weighted Dynamic Feature Fusion for Multiblock Regression (CW-DFF-MBR): A Generalized Framework for Advanced Chemometric Modelling","authors":"Meryem Nini,&nbsp;Mohamed Nohair","doi":"10.1002/cem.70118","DOIUrl":"https://doi.org/10.1002/cem.70118","url":null,"abstract":"<div>\u0000 \u0000 <p>The increasing availability of multivariate analytical data requires modelling strategies that can integrate complementary sources of information while maintaining interpretability and robustness. In this work, we propose a structured chemometric framework, termed Correlation-Weighted Dynamic Feature Fusion for Multiblock Regression (CW-DFF-MBR), to support multiblock regression by transparently integrating complementary data blocks. Rather than introducing a new learning algorithm, the proposed approach organizes established chemometric operations into a coherent workflow consisting of four steps: (i) correlation-guided variable screening, (ii) modelling of interblock interaction features, (iii) dimensionality control using principal component analysis (PCA), and (iv) correlation-weighted block fusion followed by partial least squares (PLS) regression. The methodology is first evaluated on a near-infrared (NIR) dataset of cassava roots for predicting total <i>β</i>-carotene content, which is characterized by strong collinearity and distributed spectral information. An ablation analysis is performed to examine the contribution of the different processing steps to model stability and prediction performance. The framework is then applied to a second dataset comprising wheat–flour spectra acquired under different experimental conditions to assess robustness across datasets. Results show that the proposed workflow provides stable predictive performance and consistent model interpretation across both datasets. The approach does not necessarily outperform simpler models when predictive information is concentrated in a single spectral region. However, it offers a structured, interpretable strategy for handling information distributed across correlated spectral domains. These results suggest that CW-DFF-MBR can serve as a practical framework for multiblock chemometric modelling of complex analytical datasets.</p>\u0000 </div>","PeriodicalId":15274,"journal":{"name":"Journal of Chemometrics","volume":"40 4","pages":""},"PeriodicalIF":2.1,"publicationDate":"2026-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147567989","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Augmenting Near-Infrared Spectral Data With Auxiliary Regression Generative Adversarial Network for Enhanced Prediction of Watermelon Soluble Solids Content","authors":"Xiong Li,&nbsp;Le Zhao,&nbsp;Jianhong Lai,&nbsp;Zengping Chen,&nbsp;Jiakun Su,&nbsp;Shaohe Jiao,&nbsp;Junwei Guo","doi":"10.1002/cem.70120","DOIUrl":"https://doi.org/10.1002/cem.70120","url":null,"abstract":"<div>\u0000 \u0000 <p>This study addresses the critical challenge of small sample sizes in deep learning-based detection of watermelon soluble solids content (SSC) using near-infrared spectroscopy (NIRS). We propose an auxiliary regression generative adversarial network (ARGAN) that overcomes the limitations of conventional deep convolutional generative adversarial networks (DCGAN) in regression tasks, particularly its poor controllability over chemical properties. The ARGAN framework introduces three key innovations: (1) a semisupervised architecture incorporating continuous SSC labels as conditional inputs to the generator, (2) a discriminative linear activation layer for direct SSC regression, and (3) a mean squared error-based loss function to enforce chemical-property correlations. The spectral similarity and SSC distribution consistency generated by ARGAN was higher compared to DCGAN through box-and-line plot analysis, average spectral visualization, structural similarity index (SSIMmax = 0.993), maximum mean difference (MMDmin = 0.002), and mahalanobis distance (MDmin = 1.332) analysis. More importantly, ARGAN-enhanced data improve the prediction performance of partial least squares regression (PLSR), support vector machine (SVM), and convolutional neural network (CNN) models by 9%, 9%, and 13%, respectively, with a significant reduction of 48% in the root mean square error (RMSE) of CNN. As the first successful application of ARGAN for NIRS regression enhancement, this work establishes a novel paradigm for chemically attribute-controllable data augmentation in agricultural quality detection, effectively addressing small-sample learning challenges in food spectroscopy analysis.</p>\u0000 </div>","PeriodicalId":15274,"journal":{"name":"Journal of Chemometrics","volume":"40 4","pages":""},"PeriodicalIF":2.1,"publicationDate":"2026-03-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147567991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Rapid Adulteration Detection of Fritillariae Thunbergii Bulbus: An Effective and Nondestructive Strategy Using Intelligent Algorithm-Assisted Digital Image and FT-NIR Spectroscopy","authors":"Daixin Yu,&nbsp;Qingrong Zhao,&nbsp;Tingting Lan,&nbsp;Can Zhou,&nbsp;Xuemei Cheng,&nbsp;Qiuyun Liu,&nbsp;Caiyan Dai,&nbsp;Qinan Wu,&nbsp;Cheng Qu","doi":"10.1002/cem.70117","DOIUrl":"10.1002/cem.70117","url":null,"abstract":"<div>\u0000 \u0000 <p>Fritillariae Thunbergii Bulbus (FTB) is a widely-used herb with significant nutritional and medicinal value. However, the increasing prevalence of adulterated FTB powder (FTBP) in commercial markets has compromised product quality and hindered sustainable industry development. This study developed an integrated strategy combining digital image (DI) analysis and Fourier transform near-infrared (FT-NIR) spectroscopy with intelligent algorithms to detect and quantify FTBP adulterants. The image and spectral data of FTB were used to establish classification models using single and fused datasets by traditional pattern recognition methods, machine learning, and deep learning algorithms. The commercial FTBs were used to validate the developed models. Quantitative regression models were developed using partial least squares (PLS) to predict the concentrations of adulterants in FTB. Traditional chemometrics revealed that DI and NIR dataset could initially distinguish FTBP and its adulterants. Particle swarm optimization-convolutional neural network (PSO-CNN) algorithms demonstrated superior performance by feature-level data fusion (F-LDF), achieving accuracy of 100%. External validation confirmed perfect discrimination of commercial FTBP using DI and NIR data, with predictive rates of 100%. For quantitative analysis, PLS regression yielded exceptional prediction performance, with the ratio of prediction to deviation values reaching 19.69, 7.37, and 6.09 for corn starch, soybean flour, and wheat flour adulteration using F-LDF data. This study established a rapid, nondestructive, and reliable strategy for FTB authentication, with potential applications in quality control of other herbs and spices.</p>\u0000 </div>","PeriodicalId":15274,"journal":{"name":"Journal of Chemometrics","volume":"40 3","pages":""},"PeriodicalIF":2.1,"publicationDate":"2026-03-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147565881","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Partial Least Squares Weights and Four Types of Loadings","authors":"Richard G. Brereton","doi":"10.1002/cem.70068","DOIUrl":"10.1002/cem.70068","url":null,"abstract":"&lt;p&gt;The concept of loadings when using PLS is considerably more complex than for PCA. Unfortunately, different concepts have been called by the same names, causing confusion in the literature.&lt;/p&gt;&lt;p&gt;All common PLS algorithms relate to one of these two algorithms (one with orthogonal scores and the other with nonorthogonal scores), but the loadings and scores differ in scale between methods. Due to space constraints, we will restrict our description to just these two algorithms, but all common PLS algorithms belong to one of these two classes.&lt;/p&gt;&lt;p&gt;We will now illustrate this with the data of Table 1 of the previous article [&lt;span&gt;2&lt;/span&gt;]. For brevity, we will only model variable &lt;i&gt;c&lt;/i&gt;&lt;sub&gt;1&lt;/sub&gt;—the first column of the &lt;i&gt;c&lt;/i&gt; block. Major aims of the last article were to show that there are different PLS models according to &lt;i&gt;c&lt;/i&gt; values, and that for the Martens algorithm, one can rotate the scores through different angles according to whether we use PCA or PLS, and that the rotation angle differs for each &lt;i&gt;c&lt;/i&gt; value, when there is more than one &lt;i&gt;c&lt;/i&gt; (concentration or property) value characterising a dataset.&lt;/p&gt;&lt;p&gt;This is best illustrated by a multivariate example as in Table 2, where &lt;b&gt;&lt;i&gt;X&lt;/i&gt;&lt;/b&gt; is a 15 × 5 matrix, with a maximum of 5 PLS components. The features discussed above are valid no matter how many components in the model, and can be checked by readers via the numerical data of Table 3. Data are column centred before performing PLS.&lt;/p&gt;&lt;p&gt;It can be seen that, although there are differences between x loadings, using both algorithms and the weights, and for the Martens algorithm, they also differ according to the number of components in the model, they are not enormous. Weights or loadings are commonly used in chemometrics to determine the significance of variables in a model. For example, which variables are more important markers for a metabolic process? Usually, the ones with the highest magnitude are considered most important, so they can give a clue as to which metabolites or spectral peaks are indicative of the process of interest. We will discuss the interpretation of multivariate PLS models in later articles.&lt;/p&gt;&lt;p&gt;The Wold algorithm is also commonly called the NIPALS algorithm and is very widespread in chemometrics and usually the default. However, it is always worth checking any software to ensure this is the method employed. The advantage of the Martens algorithm is that the data in scores space is a rotation of the original data, unlike the Wold algorithm.&lt;/p&gt;&lt;p&gt;As we will see in the next article, if used for estimation, both algorithms provide identical answers, so whether this difference is important depends in part on the purpose of PLS. If it is for estimation, it makes no difference, which is used, but if it is to determine the significance of variables, the different algorithms may come to different conclusions, although they will not normally be large if using the x loadings.&lt;/p&gt;&lt;p&gt;Data sharing n","PeriodicalId":15274,"journal":{"name":"Journal of Chemometrics","volume":"40 3","pages":""},"PeriodicalIF":2.1,"publicationDate":"2026-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/epdf/10.1002/cem.70068","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147564932","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Transfer Learning Neural Networks for Nuclear Forensic Image Morphology Using Image Splitting Techniques","authors":"Niko A. Petrocelli,&nbsp;Lee C. Lambert,&nbsp;Brett J. Borghetti,&nbsp;Abigail A. Bickley","doi":"10.1002/cem.70112","DOIUrl":"10.1002/cem.70112","url":null,"abstract":"<p>Manual morphological analysis of actinide particles from scanning electron microscope (SEM) imagery is a critical component of nuclear forensics but is prone to significant inter-analyst variability. To address this challenge, this work develops and evaluates an automated classification method using deep learning. We introduce a methodology based on partitioning 1906 SEM images, representing 13 classes of uranium compounds, into smaller patches for analysis. Three convolutional neural network (CNN) architectures of increasing complexity were compared: a custom baseline CNN, a simple transfer learning model using ResNet50v1, and a complex model featuring hierarchical feature extraction and a spatial attention mechanism built upon ResNet50v2. The final image classification was determined using a patch-based, class-balanced voting system. The complex model achieved a superior balanced accuracy of 94% on the test set, significantly outperforming the simple transfer learning model (89%) and the baseline model (77%). These results demonstrate that a sophisticated transfer learning architecture can serve as a robust, objective tool to increase the accuracy and consistency of actinide particle classification, thereby enhancing nuclear forensic capabilities.</p>","PeriodicalId":15274,"journal":{"name":"Journal of Chemometrics","volume":"40 3","pages":""},"PeriodicalIF":2.1,"publicationDate":"2026-03-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/epdf/10.1002/cem.70112","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147564226","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Demystifying Piecewise and Localized Scatter Correction Methods","authors":"Cannon Giglio,&nbsp;Jean-Michel Roger,&nbsp;Erik Andries","doi":"10.1002/cem.70113","DOIUrl":"10.1002/cem.70113","url":null,"abstract":"<p>Multiplicative scatter is a common source of noise in near-infrared spectroscopy and other related instrumental techniques. A wide variety of methods are commonly used for the correction of multiplicative scatter. However, the majority of such methods assume that the parameters that describe the scatter are constant throughout the measured spectrum, which is often not the case. This work investigates a family of methods that perform scatter correction using local regions of neighboring wavelengths in order to better account for wavelength-dependent scattering. The methods in question are piecewise standard normal variate, localized standard normal variate, piecewise multiplicative scatter correction, and localized multiplicative scatter correction. This work describes the theoretical and algorithmic foundations of the family of local region-based scatter correction methods and compares their application and optimization at a qualitative and quantitative level using several datasets.</p>","PeriodicalId":15274,"journal":{"name":"Journal of Chemometrics","volume":"40 3","pages":""},"PeriodicalIF":2.1,"publicationDate":"2026-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/epdf/10.1002/cem.70113","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147563337","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"化学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}