Journal of Chemometrics最新文献

This paper honors Paul Geladi with a new chemometric data visualization approach based on immersive analytics principles using virtual reality (VR). With the many technological advancements improving the accessibility of extended reality (XR), including VR, the frontier of immersive analytics is wide open for utilizing VR as a powerful chemometric data analysis tool. Immersion of human senses into a virtually generated three-dimensional (3D) world compels us to act more instinctively in complex data decision-making scenarios by using our inherent cognitive pattern recognition capacity learned over years of experience. Proposed is an immersive analytics application of VR for a hybrid human/computer-aided outlier detection process. In this application, each training set sample is virtually realized as a glyph to visually assess the inter- and intra-sample relationships present in datasets while mining for nonrepresentative samples. Sample glyph shape and size are developed using hundreds of sample similarity measurements for each sample based both on spectral and prediction property information. These sample glyphs are visually and spatially compared with one another in VR by the user. Outlier checking in VR safeguards against masking and swamping problems that are difficult to recognize with automatic algorithms. Results from dataset situations based on near infrared (NIR) and ultraviolet (UV) spectra and analyte reference values show the viability of using VR for data analysis and outlier detection. This VR application demonstrates the looming evolution of immersive analytics with an XR interface involving human reasoning in difficult chemometric data analysis settings.

Immersive analytics (IA) leverages virtual reality, augmented reality, and mixed reality to transform how users interact with complex datasets across domains such as science, industry, and education. These immersive technologies offer spatial and multimodal environments that foster intuitive exploration, but they also introduce challenges related to cognitive load, interface design, and system performance. This article presents a comprehensive review of visualization techniques, interaction models, and multimodal inputs utilized in IA. Drawing on case studies in scientific visualization, industrial training, and educational communication, we examine both the potential and limitations of current systems. Finally, we propose future research directions, focusing on real-time collaboration, adaptive user interfaces, and scalable data exploration strategies to advance the field.

Chemometrics encompasses numerous facets such as experimental design, data collection and analysis, and many others. This paper, in honor of Paul Geladi, provides our perspective on growing the scientific intuition of multivariate analysis in conventional chemometric directions not generally practicing multivariate principles. The motivation for this perspective is to express our opinion on the need for chemometrics to expand the role of the Rashomon effect beyond “many models predict well” by integrating a more comprehensive consideration of the multivariate nature of matrix effects. Described are multiple chemometric techniques that have already been enhanced by broadening the application the Rashomon effect including model selection and explanation, figures of merit (FOM), sample similarity assessment for model reliability, outlier detection, and classification—all recent research topics from the authors. This expository discussion revolves around spectroscopic data such as near infrared and fluorescence, but the concepts are relevant to other chemometric data structures.

Clustering of binary data is central to various applications, particularly in the fields of medical diagnostics, chemistry, and chemoinformatics. However, standard similarity measures often fail to capture the informative value of rare features and matching absences, treating all attributes as equally relevant. This can lead to suboptimal clustering, especially when informative patterns are hidden in low-frequency features. This study proposes a probability-weighted approach to measuring similarity, which gives more weight to rare features and accounts for the value of shared absences based on their occurrence probabilities. We analyze how this adjustment impacts clustering results, using visual comparisons and experiments on real datasets. The results show consistent gains in clustering precision and stability compared to standard measures. Our findings suggest that incorporating the rarity of features into similarity computation can offer a more reliable basis for clustering binary data, especially in domains where rare signals carry meaningful information.

The prediction results from Partial Least Squares (PLS) model are commonly used to assess whether a product meets quality standards, or whether adjustments are needed in production process parameters. It's easy to understand that misgrading is mostly occurred for marginal samples (samples near the threshold). We propose Logistic-Enhanced PLS (LE-PLS) model, which defines a logistic loss function and minimizes it via gradient descent to optimize the PLS projection vector. The prediction result of LE-PLS for marginal samples tends to be far away from the threshold value. This optimization enables LE-PLS to enhance grading capability while largely maintaining the regression accuracy of the PLS. LE-PLS was evaluated on two real-world datasets (bean pulp and corn gluten meal) and one simulated dataset, correcting 10 out of 19 misgraded samples, 6 out of 7, and 6 out of 12, respectively. Statistical analysis using paired t-tests confirmed that these improvements were significant. Although RMSEP increased slightly, the change remained within an acceptable range considering the substantial enhancement in grading performance. The algorithm has a computational complexity of $��O�\left(�\text{iteration}�*�\text{samples}�*�\text{variables}�\right)�$ during modeling training. However, its prediction-phase complexity is only $��O�\left(�\text{samples}�*�\text{variables}�\right)�$. Given these advantages, LE-PLS is well-suited for practical applications in NIR-based quality grading of products.

As a crucial extraction process in traditional Chinese medicine, quality control of percolation still faces challenges in real-time monitoring methods. To address this challenge, this study focused on the Astragalus percolation process and established an NIRS-based method for synchronous online monitoring of two bioactive markers in Astragalus percolates: Astragalus polysaccharides (APSs) and calycosin-7-O-β-D-glucoside (CG), achieving rapid and nondestructive analysis. In this study, near-infrared (NIR) spectra were collected online at different time points during percolation to determine APS and CG concentrations by means of NIRS technology, with high-performance liquid chromatography (HPLC) and ultraviolet–visible spectrophotometry (UV–Vis) used as reference methods. Two modeling approaches—partial least squares regression (PLSR) and support vector regression (SVR)—were employed to establish quantitative analytical models for these bioactive components, with model performance optimized through spectral preprocessing and feature variable selection. Results demonstrated that SVR-based models achieved superior predictive accuracy compared with PLSR. The optimal APS model showed calibration and validation set R² values of 0.9995 and 0.9874, respectively, while the CG model yielded 0.9811 (calibration) and 0.9632 (validation). Both components exhibited residual prediction deviation (RPD) values exceeding the threshold (RPD > 3), with 6.5349 for APS and 3.8357 for CG, confirming excellent predictive capability. Paired t-test analysis of external test sets (p > 0.05) revealed no statistically significant difference between measured and predicted values, further validating the model's robustness for unknown sample prediction. The concentrations of APS and CG in the Astragalus percolation solution can be simultaneously determined by this method within 30 s, significantly improving analytical efficiency compared with the conventional method (60–80 min per sample), while featuring simple operation, solvent-free consumption, low cost, and pollution-free advantages. This study demonstrates that the combination of NIRS and chemometrics enables real-time monitoring of multiple key substance concentrations during the percolation process. As a green analytical technology, NIRS shows significant potential for improving production efficiency and ensuring product quality consistency.

This computational study aimed to demonstrate distinct characteristics of alpha-D-mannopyranoside structure, leveraging D-mannose and its analogs due to their known roles in host–pathogen interactions and potential to be used as nutraceuticals. Targeting bacterial adhesion is a critical strategy to combat urinary tract infections (UTIs), especially given rising antibiotic resistance. The FimH lectin on Escherichia coli is a key mediator of this adhesion, making it a compelling target for novel anti-adhesive therapies. We employed a multi-stage virtual screening pipeline to efficiently explore a vast chemical space around the ligands and their binding interactions. Ligand-based virtual screening, utilizing self-organizing maps (SOMs), clustered 5256 D-mannose-similar structures, identifying a promising subset of 141 molecules with 39 known bioassay actives. This was followed by structure-based ligand docking to precisely evaluate their inhibitory impact on FimH lectin. To understand the structural features driving activity, principal component analysis (PCA) was then applied to analyze the molecular structures and their physicochemical descriptors. Our analysis revealed that 15 compounds exhibited the highest binding energy and docking scores. Crucially, the alpha-D-mannopyranoside conformation demonstrated the most effective inhibitory profile. This superior activity, despite structural similarities, was differentiated by two 3D-matrix descriptors: HRG and Wi G, highlighting their significance in predicting subtle yet impactful conformational preferences.

To address the issue of insufficient fault detection performance of global–local preserving projections (GLPP) in the detection of minor faults within nonlinear dynamic processes, a novel fault detection method based on GLPP and Kantorovich distance combined with a sliding window (GLPP-KD) is proposed. Firstly, the GLPP algorithm is used to construct a weight matrix to retain the key information of the data, and the objective function containing local and global information is transformed into a generalized eigenvector problem to obtain a projection matrix. Additionally, the sliding window technique integrated with the Kantorovich distance is employed to quantify the discrepancies between probability distributions, thereby capturing the local dynamic characteristics of the data. Eventually, the fault detection task is achieved by identifying the minor distinctions between normal and faulty states. Experimental results show that compared with traditional methods, GLPP-KD improves the fault detection accuracy and effectively reduces the false alarm rate. The proposed method provides a strong guarantee for the safe and stable operation of the industry and has high application value.