Search Results (502)

Impulsive overtopping generated by dam-break surges is a critical hazard for dikes and flood-protection embankments, especially in reservoirs and mountainous catchments. Unlike classical coastal wave overtopping, which is governed by long, irregular wave trains and usually characterized by mean overtopping discharge over many waves, these dam-break-type events are dominated by one or a few strongly nonlinear bores with highly transient overtopping heights. Accurately predicting the resulting overtopping levels under such impulsive flows is therefore important for flood-risk assessment and emergency planning. Conventional cluster-then-predict approaches, which have been proposed in recent years, often first partition data into subgroups and then train separate models for each cluster. However, these methods often suffer from rigid boundaries and ignore the uncertainty information contained in clustering results. To overcome these limitations, we propose a GMM+MoE framework that integrates Gaussian Mixture Model (GMM) soft clustering with a Mixture-of-Experts (MoE) predictor. GMM provides posterior probabilities of regime membership, which are used by the MoE gating mechanism to adaptively assign expert models. Using SPH-simulated overtopping data with physically interpretable dimensionless parameters, the framework is benchmarked against XGBoost, GMM+XGBoost, MoE, and Random Forest. Results show that GMM+MoE achieves the highest accuracy ($R^2=0.9638$ on the testing dataset) and the most centralized residual distribution, confirming its robustness. Furthermore, SHAP-based feature attribution reveals that relative propagation distance and wave height are the dominant drivers of overtopping, providing physically consistent explanations. This demonstrates that combining soft clustering with adaptive expert allocation not only improves accuracy but also enhances interpretability, offering a practical tool for dike safety assessment and flood-risk management in reservoirs and mountain river valleys. Full article

This study provides a detailed comparative analysis of a three-hybrid intrusion detection method aimed at strengthening network security through precise and adaptive threat identification. The proposed framework integrates an Autoencoder-Gaussian Mixture Model (AE-GMM) with two supervised learning techniques, XGBoost and Logistic Regression, combining deep feature extraction with interpretability and stable generalization. Although the downstream classifiers are trained in a supervised manner, the hybrid intrusion detection nature of the framework is preserved through unsupervised representation learning and probabilistic modeling in the AE-GMM stage. Two benchmark datasets were used for evaluation: NSL-KDD, representing traditional network behavior, and UNSW-NB15, reflecting modern and diverse traffic patterns. A consistent preprocessing pipeline was applied, including normalization, feature selection, and dimensionality reduction, to ensure fair comparison and efficient training. The experimental findings show that hybridizing deep learning with gradient-boosted and linear classifiers markedly enhances detection performance and resilience. The AE–GMM-XGBoost model achieved superior outcomes, reaching an F1-score above 0.94 ± 0.0021 and an AUC greater than 0.97 on both datasets, demonstrating high accuracy in distinguishing legitimate and malicious traffic. AE-GMM-Logistic Regression also achieved strong and balanced performance, recording an F1-score exceeding 0.91 ± 0.0020 with stable generalization across test conditions. Conversely, the standalone AE-GMM effectively captured deep latent patterns but exhibited lower recall, indicating limited sensitivity to subtle or emerging attacks. These results collectively confirm that integrating autoencoder-based representation learning with advanced supervised models significantly improves intrusion detection in complex network settings. The proposed framework therefore provides a solid and extensible basis for future research in explainable and federated intrusion detection, supporting the development of adaptive and proactive cybersecurity defenses. Full article

High-dimensional regression with multivariate responses poses significant challenges when data are collected across multiple platforms, each with potentially correlated outcomes. In this paper, we introduce a multi-platform multivariate high-dimensional linear regression (MM-HLR) model for simultaneously modeling within-platform correlation and cross-platform information fusion. Our approach incorporates a mixture of Lasso and group Lasso penalties to promote both individual predictor sparsity and cross-platform group sparsity, thereby enhancing interpretability and estimation stability. We develop an efficient computational algorithm based on iteratively reweighted least squares and block coordinate descent to solve the resulting regularized optimization problem. We establish theoretical guarantees for our estimator, including oracle bounds on prediction error, estimation accuracy, and support recovery under mild conditions. Our simulation studies confirm the method’s strong empirical performance, demonstrating low bias, small variance, and robustness across various dimensions. The analysis of real financial data further validates the performance gains achieved by incorporating multivariate responses and integrating data across multiple platforms. Full article

Identified-hadron production (p, ${\pi }^{\pm }$, $K^{\pm }$) in ${\pi }^{+}+\mathrm{Be}$ at $p_{\mathrm{lab}}=60\mathrm{GeV}/c$ ($\sqrt{s}\simeq 10.6\mathrm{GeV}$) is investigated using Pythia 8.315 (Monash tune) with the Angantyr extension. Differential multiplicities $d^{2}n/\left(dpd\theta \right)$ are confronted with NA61/SHINE measurements across standard $\theta$ bins. Within the fluctuating-radii Double-Strikman (DS) scheme, two unsuppressed opacity mappings are compared to quantify systematics. In addition, a minimal extension is introduced: a flat, post-classification, channel-wise acceptance applied after ND/SD/DD/EL tagging. It acts on primary and secondary $\pi N$ pairs, keeps hadronization fixed (Lund string), and leaves the internal event generation of each admitted subcollision unchanged. Opacity-mapping variations alone induce only percent-level differences and do not resolve the soft/forward tensions. By contrast, the flat acceptance—interpretable as a reduced effective ND weight—improves agreement across species and angles. It hardens the forward ${\pi }^{+}$ spectra and lowers large-$\theta$ yields, produces milder charge-asymmetric changes for ${\pi }^{-}$ consistent with the weaker leading feed, suppresses proton yields at all angles (with a residual $\sim 30\%$ forward high-p deficit), and improves $K^{\pm }$, with a stronger effect for $K^{+}$ than $K^{-}$. These results show that a geometry-blind reweighting of the subcollision mixture suffices to capture the main NA61/SHINE trends for ${\pi }^{+}+\mathrm{Be}$ at SPS energies without modifying hadronization. The approach provides a controlled baseline for subsequent, channel-balanced refinements and broader ${\pi }^{+}A$ tuning. Full article

Accurate short-term forecasting of solar photovoltaic (PV) power output is essential for efficient grid integration and energy management, especially given the widespread global adoption of PV systems. To address this research gap, the present study introduces a scalable, interpretable ensemble learning model of PV power prediction with respect to a large PVOD v1.0 dataset, which encompasses more than 270,000 points representing ten PV stations. The proposed methodology involves data preprocessing, feature engineering, and a hybrid ensemble model consisting of Random Forest, XGBoost, and CatBoost. Temporal features, which included hour, day, and month, were created to reflect the diurnal and seasonal characteristics, whereas feature importance analysis identified global irradiance, temperature, and temporal indices as key indicators. The hybrid ensemble model presented has a high predictive power, with an R² = 0.993, a Mean Absolute Error (MAE) = 0.227 kW, and a Root Mean Squared Error (RMSE) = 0.628 kW when applied to the PVOD v1.0 dataset to predict short-term PV power. These findings were achieved on standardized, multi-station, open access data and thus are not in an entirely rigorous sense comparable to previous studies that may have used other datasets, forecasting horizons, or feature sets. Rather than asserting numerical dominance over other approaches, this paper focuses on the real utility of integrating well-known tree-based ensemble techniques with time-related feature engineering to derive real, interpretable, and computationally efficient PV power prediction models that can be used in smart grid applications. This paper shows that a mixture of conventional ensemble methods and extensive temporal feature engineering is effective in producing consistent accuracy in PV forecasting. The framework can be reproduced and run efficiently, which makes it applicable in the integration of smart grid applications. Full article

This study investigates the relationship between apparent electrical resistivity (ER) and key material parameters governing moisture and pore-structure characteristics of concrete. An experimental program was conducted using six concrete mix designs, where ER was continuously measured under controlled wetting and drying cycles to characterize its dependence on the degree of saturation (DS). Results confirmed that ER decreases exponentially with increasing DS across all mixtures, with R² values between 0.896 and 0.997, establishing DS as the dominant factor affecting electrical conduction. To incorporate additional pore-structure parameters, eight input combinations consisting of DS, porosity (P), water–cement ratio (WCR), and compressive strength (f′c) were evaluated using five machine learning models. Gaussian Process Regression and Neural Networks achieved the highest accuracy, particularly when all parameters were included. SHAP analysis revealed that DS accounts for the majority of predictive influence, while porosity and WCR provide secondary but meaningful contributions to ER behavior. Guided by these insights, nonlinear multivariate regression models were formulated, with the exponential model yielding the strongest predictive capability (R² = 0.96). The integrated experimental–computational approach demonstrates that ER is governed by moisture dynamics and pore-structure refinement, offering a physically interpretable and statistically robust framework for nondestructive durability assessment of concrete. Full article

Secchi disk depth (SDD) is a widely critical indicator of water transparency. However, existing retrieval models often suffer from limited transferability and biased predictions when applied to optically diverse waters. Here, we compiled a dataset of 6218 paired in situ SDD and remote sensing reflectance (R_rs) measurements to evaluate model generalization. We benchmarked nine machine learning (ML) models (RF, KNN, SVM, XGB, LGBM, CAT, RealMLP, BNN-MCD, and MDN) under three validation scenarios with progressively decreasing training-test overlap: Random, Waterbody, and Cross-Optical Water Type (Cross-OWT). Furthermore, SHAP analysis was employed to interpret feature contributions and relate model behaviors to optical properties. Results revealed a distinct scenario-dependent generalization gradient. Random splits yielded minimal bias. In contrast, Waterbody transfer consistently shifted predictions toward underestimation (SSPB: −16.9% to −3.8%). Notably, Cross-OWT extrapolation caused significant error inflation and a bias reversal toward overestimation (SSPB: 10.7% to 88.6%). Among all models, the Mixture Density Network (MDN) demonstrated superior robustness with the lowest overestimation (SSPB = 10.7%) under the Cross-OWT scenario. SHAP interpretation indicated that engineered indices, particularly NSMI, functioned as regime separators, with substantial shifts in feature attribution occurring at NSMI values between 0.4 and 0.6. Accordingly, feature sensitivity analysis showed that removing band ratios and indices improved Cross-OWT robustness for several classical ML models. For instance, KNN exhibited a significant reduction in Median Symmetric Accuracy (MdSA) from 96% to 40% after feature reduction. These findings highlight that model applicability must be evaluated under scenario-specific conditions, and feature engineering strategies require rigorous testing against optical regime shifts to ensure generalization. Full article

This paper introduces a novel extension of the classical Lindley distribution, termed the X-Lindley model, obtained by a specific mixture of exponential and Lindley distributions, thereby substantially enriching the distributional flexibility. To enhance its inferential scope, a comprehensive reliability analysis is developed under a generalized progressive hybrid censoring scheme, which unifies and extends several traditional censoring mechanisms and allows practitioners to accommodate stringent experimental and cost constraints commonly encountered in reliability and life-testing studies. Within this unified censoring framework, likelihood-based estimation procedures for the model parameters and key reliability characteristics are derived. Fisher information is obtained, enabling the establishment of asymptotic properties of the frequentist estimators, including consistency and normality. A Bayesian inferential paradigm using Markov chain Monte Carlo techniques is proposed by assigning a conjugate gamma prior to the model parameter under the squared error loss, yielding point estimates, highest posterior density credible intervals, and posterior reliability summaries with enhanced interpretability. Extensive Monte Carlo simulations, conducted under a broad range of censoring configurations and assessed using four precision-based performance criteria, demonstrate the stability and efficiency of the proposed estimators. The results reveal low bias, reduced mean squared error, and shorter interval lengths for the XLindley parameter estimates, while maintaining accurate coverage probabilities. The practical relevance of the proposed methodology is further illustrated through two real-life data applications from engineering and physical sciences, where the XLindley model provides a markedly improved fit and more realistic reliability assessment. By integrating an innovative lifetime model with a highly flexible censoring strategy and a dual frequentist–Bayesian inferential framework, this study offers a substantive contribution to modern survival theory. Full article

NMR spectroscopy enables the study of complex mixtures, including plant extracts. The interpretation of specific ranges of ¹H NMR spectra allows for the determination of polyphenolic compound, sugar, amino acid, and fatty acid profiles. The main goal of ¹H NMR analyses of plant extracts is to identify the unique “fingerprint” of the material being studied. The aim of this study was to determine the metabolomic profile and antioxidant activity of various Lavandula angustifolia (Betty’s Blue, Elizabeth, Hidcote, and Blue Mountain White) and Lavandula × intermedia cultivars (Alba, Grosso, and Gros Bleu) grown in Poland. Modern green chemistry extraction methods (supercritical fluid extraction (SFE) and ultrasound-assisted extraction (UAE)) were used to prepare the lipophilic and hydrophilic extracts, respectively. The secondary metabolite profiles were determined using the diagnostic signals from ¹H NMR and HPLC-DAD analyses. These metabolomic profiles were used to illustrate the differences between the different lavender and lavandin cultivars. The HPLC-DAD analysis revealed that both lavender species have similar polyphenolic profiles but different levels of individual compounds. The extracts from L. angustifolia were characterized by higher phenolic acid and flavonoid contents, while the extracts from L. × intermedia had a higher coumarin content. Diagnostic ¹H NMR signals can be used to verify the authenticity and origin of plant extracts, and identify directions for further research, providing a basis for applications such as in cosmetics. Full article

Background: Ulcerative colitis (UC) is characterized by chronic mucosal inflammation, oxidative stress, and disruption of intestinal metabolic homeostasis. Immunomodulatory nutrients such as arginine, glutamine, and β-hydroxy β-methylbutyrate (HMB) have shown potential benefits; however, their combined molecular effects on UC remain insufficiently defined. Objective: To investigate the individual and combined effects of arginine, glutamine, and HMB on inflammatory and metabolic gene expression, oxidative stress markers, and histopathological outcomes in a dextran sulfate sodium (DSS)-induced colitis model. Methods: Female Sprague Dawley rats were assigned to six groups: control, DSS, DSS + arginine, DSS + glutamine, DSS + HMB, and DSS + mixture. Colitis was induced using 3% DSS. Colon tissues were examined histologically, serum MDA, MPO, and GSH levels were quantified, and mRNA expression of IL6, IL10, COX2, NOS2, ARG2, CCR1, and ALDH4A1 was measured by RT-qPCR. Pathway enrichment analyses were performed to interpret cytokine and metabolic network regulation. Results: DSS induced severe mucosal injury, elevated MDA and MPO, reduced GSH, and significantly increased IL6, COX2, NOS2, ARG2, and CCR1 expression. Glutamine demonstrated the strongest anti-inflammatory and antioxidant effects by decreasing IL6 and COX2 and restoring GSH. Arginine primarily modulated nitric oxide–related pathways, whereas HMB increased ALDH4A1 expression and metabolic adaptation. The combination treatment produced more balanced modulation across inflammatory, chemokine, and metabolic pathways, consistent with enrichment results highlighting cytokine signaling and amino acid metabolism. Histopathological improvement was greatest in the mixture group. Conclusions: Arginine, glutamine, and HMB ameliorate DSS-induced colitis through coordinated regulation of cytokine networks, oxidative stress responses, and metabolic pathways. Their combined use yields broader and more harmonized therapeutic effects than individual administration, supporting their potential as targeted immunonutritional strategies for UC. Rather than targeting a single inflammatory mediator, this study was designed to test whether combined immunonutrient supplementation could promote coordinated regulation of cytokine signaling, oxidative stress responses, and metabolic adaptation, thereby facilitating mucosal repair in experimental colitis. Full article

In machine learning, the Bayes classifier represents the theoretical optimum for minimizing classification errors. Since estimating high-dimensional probability densities is impractical, simplified approximations such as naïve Bayes and k-nearest neighbor are widely used as baseline classifiers. Despite their simplicity, these methods require design choices—such as the distance measures in kNN, or the feature independence in naïve Bayes. In particular, naïve Bayes relies on implicit assumptions by using Gaussian mixtures or univariate kernel density estimators. Such design choices, however, often fail to capture heterogeneous distributional structures across features. We propose a flexible naïve Bayes classifier that leverages Pareto Density Estimation (PDE), a parameter-free, non-parametric approach shown to outperform standard kernel methods in exploratory statistics. PDE avoids prior distributional assumptions and supports interpretability through visualization of class-conditional likelihoods. In addition, we address a recently described pitfall of Bayes’ theorem: the misclassification of observations with low evidence. Building on the concept of plausible Bayes, we introduce a safeguard to handle uncertain cases more reliably. While not aiming to surpass state-of-the-art classifiers, our results show that PDE-flexible naïve Bayes with uncertainty handling provides a robust, scalable, and interpretable baseline that can be applied across diverse data scenarios. Full article

In medical diagnostics and decision-making, particularly in dentistry where structural interpretation of radiographs plays a crucial role, accurate image segmentation is a fundamental step. One established approach to segmentation is the use of clustering techniques. This study evaluates the performance of five clustering algorithms, namely, K-Means, Fuzzy C-Means, DBSCAN, Gaussian Mixture Models (GMM), and Agglomerative Hierarchical Clustering for image segmentation. Our study uses two sets of real-world dental data comprising 140 adult tooth images and 70 children’s tooth images, including professionally annotated ground truth masks. Preprocessing involved grayscale conversion, normalization, and image downscaling to accommodate computational constraints for complex algorithms. The algorithms were accessed using a variety of metrics including Rand Index, Fowlkes-Mallows Index, Recall, Precision, F1-Score, and Jaccard Index. DBSCAN achieved the highest performance on adult data in terms of structural fidelity and cluster compactness, while Fuzzy C-Means excelled on the children dataset, capturing soft tissue boundaries more effectively. The results highlight distinct performance behaviours tied to morphological differences between adult and pediatric dental anatomy. This study offers practical insights for selecting clustering algorithms tailored to dental imaging challenges, advancing efforts in automated, label-free medical image analysis. Full article

Accurate prediction of the mechanical and rheological properties of self-compacting concrete (SCC) is critical for mixture design and engineering decision-making; however, conventional empirical approaches often struggle to capture the coupled nonlinear relationships among mixture variables. To address this challenge, this study develops an integrated and interpretable hybrid machine learning (ML) framework by coupling three ML models (RF, XGBoost, and SVR) with five metaheuristic optimizers (SSA, PSO, GWO, GA, and WOA), and by incorporating SHAP and partial dependence (PDP) analyses for explainability. Two SCC datasets with nine mixture parameters are used to predict 28-day compressive strength (CS) and slump flow (SF). The results show that SSA provides the most stable hyperparameter optimization, and the best-performing SSA–RF model achieves test R² values of 0.967 for CS and 0.958 for SF, with RMSE values of 2.295 and 23.068, respectively. Feature importance analysis indicates that the top five variables contribute more than 80% of the predictive information for both targets. Using only these dominant features, a simplified SSA–RF model reduces computation time from 7.3 s to 5.9 s and from 9.7 s to 6.1 s for the two datasets, respectively, while maintaining engineering-level prediction accuracy, and the SHAP and PDP analyses provide transparent feature-level explanations and verify that the learned relationships are physically consistent with SCC mixture-design principles, thereby increasing the reliability and practical applicability of the proposed framework. Overall, the proposed framework delivers accurate prediction, transparent interpretation, and practical guidance for SCC mixture optimization. Full article

Multimodal question answering (QA) involves integrating information from both visual and textual inputs and requires models that can reason compositionally and accurately across modalities. Existing approaches, including fine-tuned vision–language and prompting, often struggle with generalization, interpretability, and reliance on task-specific data. In this work, we propose a Mixture-of-Reasoning Agents (MiRA) framework for zero-shot multimodal reasoning. MiRA decomposes the reasoning process across three specialized agents—Visual Analyzing, Text Comprehending, and Judge—which consolidate multimodal evidence. Each agent operates independently using pretrained language models, enabling structured, interpretable reasoning without supervised training or task-specific adaptation. Evaluated on the ScienceQA benchmark, MiRA achieves 96.0% accuracy, surpassing all zero-shot methods, outperforming few-shot GPT-4o models by more than 18% on image-based questions, and achieving similar performance to the best fine-tuned systems. The analysis further shows that the Judge agent consistently improves the reliability of individual agent outputs, and that strong linear correlations (r > 0.95) exist between image-specific accuracy and overall performance across models. We identify a previously unreported and robust pattern in which performance on image-specific tasks strongly predicts overall task success. We also conduct detailed error analyses for each agent, highlighting complementary strengths and failure modes. These results demonstrate that modular agent collaboration with zero-shot reasoning provides highly accurate multimodal QA, establishing a new paradigm for zero-shot multimodal AI and offering a principled framework for future research in generalizable AI. Full article

Near-infrared (NIR) spectroscopy is a rapid, non-destructive analytical tool widely used in the food and agricultural sectors. In this study, two NIR instruments were compared for classifying the addition of microplastics (MPs) to high-moisture-content samples such as vegetables and fruit. Polyethylene (PE), polypropylene (PP), and a mix of polymers (PE + PP) MP were added to mixtures of spinach and banana and scanned using benchtop (Bruker Tango) and portable (MicroNIR) instruments. Both principal component analysis (PCA) and partial least squares (PLS) were used to analyze and interpret the spectra of the samples. Quantitative models were developed to predict the addition of Mix, PP, or PE to spinach and banana samples using PLS regression. The R² _CV and the SECV obtained were 0.88 and 0.44 for the benchtop samples, and 0.54 and 0.67 for the portable instruments, respectively. Two wavenumber regions were also evaluated: 11,520–7500 cm⁻¹ (short to medium wavelengths), and 7500–4200 cm⁻¹ (long wavelengths). The R² _CV and the SECV obtained were 0.88 and 0.46, 0.86 and 0.49, respectively, for the prediction of addition in samples analyzed on the benchtop instrument using short and long wavenumbers, respectively. This study provides new insights into the comparison of two instruments for detecting the addition of MPs in high-moisture samples. The results of this study will ensure that NIR can be utilized not only to measure the quality of these samples but also to monitor MPs. Full article