Search Results (1,883)

Normal contact stiffness of textured interfaces is controlled by the load-bearing contribution of deterministic texture and by the nonlinear response of rough load-bearing regions. This study formulates a hybrid deterministic–microlevel model that couples regular relief patterns—including dimples, grooves, periodic ribs and scraped high points—with a micromechanical representation of plateau roughness. Depending on texture topology and scale hierarchy, the microlevel response is represented either by a Greenwood–Williamson (GW) statistical contact model with a smooth elastic–plastic (EP) transition or by a fractal contact model. The deterministic level accounts for open-area fraction, texture depth and load redistribution, and it includes a finite-gauge spectral correction for periodic ribs and grooves to account for the finite measurement window. In a metallic dimple benchmark, the hybrid deterministic-texture/GW–EP formulation yields a mean relative error of 16.3% across all data points in the two selected textured series. In a saturated square-wave benchmark, the finite-gauge spectral correction yields a mean relative error of 10.37% for the independent patterned points. A compliance-based topology criterion is then established to determine, from open-area fraction, element depth, applied load and the ratio between texture period and plateau-roughness spacing, whether stiffness is governed primarily by deterministic texture, by micro-roughness or by their coupled response. The resulting formulation supports early-stage design exploration without requiring a full three-dimensional contact calculation at every parameter point. Independent periodic three-dimensional checks for circular-dimple cells showed that the finest FE solution agreed with the spectral prediction of the deterministic normal approach within 1.2%; the correction estimated by direct unilateral BEM changed total stiffness by no more than approximately 5% in the tested texture-influenced case. Full article

The release of ChatGPT in November 2022 transformed second language (L2) writing instruction and led to rapid growth in research on large language model (LLM)-generated feedback; however, no synthesis has mapped this literature in terms of feedback quality, learner uptake, and pedagogical integration. This scoping review examines 185 empirical studies published between November 2022 and March 2026 that were identified through a Scopus search (n = 283 screened) and analysed using a systematic keyword-based charting framework applied to full abstracts, with full-text analysis of 35 studies. The review identifies four major patterns: (1) comparative AI–human feedback research dominates the literature (27.6%); (2) content-level feedback remains underexplored (13.5% of studies); (3) learner uptake is rarely measured as a primary outcome; and (4) LLM feedback is broadly comparable to teacher feedback for surface-level errors but weaker for content and argumentation, while learner perceptions often exceed demonstrated performance outcomes. Hybrid AI–teacher models show promising but underexamined potential, accounting for only 12.4% of the literature. The field shows a focus on perceptions rather than learning outcomes, an apparent tendency toward positive-results reporting, and no clear teaching models. This study proposes a typology of LLM feedback functions and outlines a research agenda focused on uptake, longitudinal outcomes, and hybrid AI–teacher integration. Full article

Rapid and accurate identification of mine water inrush sources is critical for hazard control in underground coal mining. Conventional Fisher discriminant analysis is often limited by feature redundancy and multicollinearity when applied to small-sample, high-dimensional hydrochemical data. To address this, we propose GWO–Fisher, a hybrid model integrating the Grey Wolf Optimizer (GWO) with Fisher discriminant analysis. The model employs correlation-based pre-screening followed by global optimization, using a fitness function that combines Fisher accuracy with a feature-size penalty, to achieve a compact and interpretable feature set. Trained on data from the Xiegou Coal Mine (Shanxi, China), it reduced 17 hydrochemical indicators to 12 key features, achieving 92.98% training accuracy and 86.21% test accuracy—an improvement of 10.35 percentage points over conventional Fisher. When independently validated across four mines in three spring domains, the model maintained over 83% accuracy, consistently selecting TDS, K⁺, and HCO₃⁻ as core features. Misclassification patterns were cross-domain consistent and linked to hydrogeological conditions. The proposed GWO–Fisher model balances predictive accuracy with hydrogeological interpretability, demonstrating reliable performance across both single-mine and cross-spring-domain scenarios. Full article

Power quality disturbance (PQD) classifiers can lose 30–50 percentage points of accuracy when the signal-to-noise ratio (SNR) drops to 5 dB. This study finds that the degradation arises primarily not from architectural limitations but from hyperparameter configurations that fail to adapt across noise levels, within the IEEE 1159 disturbance set and additive-noise conditions studied here. Building on this insight, we propose a one-dimensional convolutional neural network (1D-CNN) with channel attention, whose three key hyperparameters—the learning rate, the first-layer convolutional kernel size, and the attention reduction ratio—are automatically optimized using the Phong Optimization Algorithm (POA). A 12-class synthetic dataset constructed to IEEE 1159-2019, with noise levels from noise-free to 5 dB SNR, is used for training and evaluation. Under the extreme condition of 5 dB SNR, the proposed method achieves 83.4% accuracy, outperforming a support vector machine (SVM) with discrete wavelet transform features (66.30%), a hybrid CNN–long short-term memory network (CNN-LSTM; 43.32%), and plain 1D-CNN (38.39%) under their commonly reported configurations. When every deep-learning baseline receives the same POA hyperparameter optimization under a fair per-SNR protocol, this advantage largely disappears: at 5 dB SNR all POA-optimized deep methods fall within roughly 5 percentage points (82.31–87.13%), and the attention module’s own contribution shrinks to within run-to-run variation (the same proposed model scoring 82.77% with attention vs. 82.48% without), showing that systematic hyperparameter optimization, not architectural novelty, drives the noise robustness. A single-set ablation under the original 10 dB-optimized configuration points the same way (POA optimization alone raising 5 dB accuracy from 41.6% to 76.4%, with channel attention adding a further 7 points). A comparison against classical threshold-index classifiers on the identical dataset shows the same pattern at the level of hand-crafted features: with fixed clean-calibrated thresholds, the index classifier collapses from 95.7% to 17.5% at 5 dB SNR, while per-noise-level re-calibration of the same indices recovers 83.8%. Furthermore, the POA-optimized hyperparameters were validated on an independent public PQD dataset, achieving 90.00% accuracy when training a fresh model from scratch. It also stays robust under more realistic complex noise (80.9% at 5 dB SNR) and, on a two-class real-measured probe, transfers to field signals with only light calibration. These findings suggest that for noise-robust PQD classification, hyperparameter optimization deserves as much attention as the architectural design itself, rather than being treated as a final tuning step. Full article

Artificial intelligence (AI) marketing systems increasingly plan campaigns, generate messages, assess performance, and revise outputs with limited human input. In such multi-agent systems (MAS), commercial optimization pressure may produce ethical compliance drift: gradual movement from acceptable persuasion toward detector-defined manipulation-risk signals. We examine this problem in a controlled OpenClaw simulation of a three-role AI marketing team. The study varies optimization pressure across baseline, moderate-pressure, and high-pressure operating regimes while holding the model backend, prompt pool, agent roles, and monitoring process constant. Optimization pressure significantly affected detector-defined manipulation-risk scores, and the observed pattern was non-monotonic. The moderate-pressure condition produced the highest observed mean, but moderate and high pressure were not statistically distinguishable in the pairwise comparison. These results reflect detector-based risk indicators in simulated marketing-agent outputs, not direct evidence of consumer harm, deception, or real-world behavioral manipulation. The findings support pressure-aware auditing as a standards-informed monitoring practice consistent with IEEE value-sensitive design principles, while also identifying the need for human annotation, hybrid detectors, and real-user validation before stronger governance claims are made. Full article

Prostate adenocarcinoma is characterized by substantial inter-patient heterogeneity, limiting the clinical reliability of conventional diagnostic tools, including prostate-specific antigen testing. This limitation underscores the need for robust molecular biomarkers that may complement conventional diagnostic tools, highlighting the urgent need for biomarkers capable of enhancing diagnostic accuracy and enabling more precise risk stratification. In the present study, transcriptomic data from The Cancer Genome Atlas (TCGA) were analyzed using an integrative bioinformatics and machine learning pipeline., The proposed workflow was designed as a stepwise and reproducible biomarker prioritization framework in which differential expression analysis, functional enrichment, protein–protein interaction (PPI) based network interpretation, graph-convolutional feature selection, and hybrid ensemble machine learning were sequentially integrated. Differential gene expression analysis was combined with pathway enrichment (Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Reactome), protein–protein interaction network construction, and graph-convolutional feature selection. Multiple machine learning algorithms, including Random Forest, Gradient Boosting Machine, Support Vector Classifier, Artificial Neural Network, and AdaBoost, were systematically evaluated. A hybrid ensemble model integrating Gradient Boosting Machine and Random Forest (GBM+RF) was subsequently developed. Model performance was assessed using accuracy, sensitivity, specificity, and area under the Receiver Operating Characteristic (ROC) and externally validated using the independent GSE14206 dataset. The analysis revealed a coordinated molecular pattern characterized by dysregulated cell cycle activity and enhanced interferon-mediated immune signaling. Protein–protein interaction analysis identified STAT1 and PLK1 as highly connected network hub genes within immune-related and cell-cycle-associated modules. Among the evaluated models, the hybrid GBM+RF framework achieved the highest predictive performance on the TCGA dataset, with AUC: 0.9526; Accuracy: 97.49%. External validation using the GSE14206 dataset confirmed the robustness of this model (AUC: 0.9156; Accuracy: 91.53%). These findings support a broader multi-gene candidate signature in prostate adenocarcinoma, in which machine learning prioritized genes such as XAF1, APP, RPA3, IFIH1, UBE2D2, RSAD2, KIF2C, and PLK1, while STAT1 and PLK1 provided complementary network-level biological relevance. The proposed framework provides a robust and transferable strategy for biomarker discovery and precision oncology. Full article

Class imbalance remains one of the most challenging problems in sentiment analysis, particularly in tourism review datasets where positive opinions substantially outnumber neutral and negative comments. This issue is especially critical because minority classes often contain the most valuable information regarding customer dissatisfaction, service failures, and opportunities for improvement. In this work, we propose a hybrid balancing methodology for sentiment analysis in Mexican Spanish tourism reviews that combines undersampling and Large Language Model (LLM)-based oversampling. The proposed framework first extracts representative words from each sentiment class using Mutual Information, then enriches them through dictionary-based or embedding-based lexical substitutions, and finally generates synthetic reviews using GPT-4o-mini guided by these representative terms. Experiments were conducted on a corpus that contains approximately 300,000 tourism reviews collected from TripAdvisor, exhibiting severe sentiment imbalance. Three undersampling strategies and multiple oversampling configurations were evaluated across six traditional machine learning classifiers and one Transformer-based model (BETO). Results show that random undersampling consistently outperformed centroid-based and K-means-based alternatives while also requiring the lowest computational cost. The best overall performance was obtained by BETO, achieving a Macro-F1 score of 0.57 compared to 0.51 on the original imbalanced dataset, representing an improvement of 11.8%. Significant gains were also observed for minority classes, with improvements exceeding 16% for the most underrepresented category. Furthermore, the proposed methodology consistently outperformed direct prompt-based generation using GPT-4o-mini, Gemini 2.5 Flash, and Llama 3.3 70B. These findings suggest that guiding synthetic review generation through representative words effectively preserves domain-specific lexical and semantic patterns of Mexican Spanish tourism reviews, resulting in more balanced datasets and improved sentiment classification performance. Full article

Artificial intelligence architecture is no longer adequately described by model size alone. Dense Transformers remain the reference architecture for language and multimodal reasoning, but production systems increasingly combine conditional computation, retrieval, memory, tools, verifiers, edge-cloud routing, observability and governance. This review makes three engineering claims. First, sparse Mixture-of-Experts models are currently the clearest capacity-scaling pattern, because they decouple total parameters from active per-token computation, although routing imbalance and distributed communication remain hard constraints. Second, state-space, recurrent and linear attention hybrids are best interpreted as attention-budgeting architectures: they reduce KV-cache and long-context costs, but do not yet displace dense attention in every reasoning regime. Third, JEPA-style latent world models change the learning objective from surface-token or pixel prediction to representation prediction, which is strategically important for perception and planning but still not a drop-in replacement for general language interfaces. To make the maturity claims auditable, this review uses a PRISMA-inspired search protocol, an explicit technology readiness rubric, quantitative comparison tables, hardware and memory-bandwidth analysis, deployment and reproducibility categories, and failure cases for RAG and agents. The main conclusion is that the optimal architecture is task- and constraint-dependent: small dense or hybrid models are often preferred for real-time edge inference, RAG and graph memory for changing enterprise knowledge, frontier dense or sparse models for difficult reasoning, and agentic workflows only when tool permissions, rollback, provenance and human oversight are engineered as first-class components. Full article

The rapid growth of Intelligent Transportation Systems (ITSs) and Internet of Things (IoT) technologies has generated massive volumes of distributed traffic data, creating significant challenges related to privacy, scalability, communication overhead, and heterogeneous traffic patterns. To address these challenges, this paper proposes FedTraffic, a hierarchical federated learning framework for traffic flow forecasting that integrates Edge–Fog–Cloud computing, hybrid deep learning, adaptive federated optimization, and Explainable Artificial Intelligence (XAI). The proposed framework combines a Temporal Convolutional Network–Conditional Variational Autoencoder (TCN–CVAE) with traffic-behavior clustering, adaptive client selection, and hierarchical model aggregation to enable accurate, privacy-preserving, and interpretable traffic prediction under heterogeneous non-IID environments. Extensive experiments demonstrate that FedTraffic achieves a best Mean Absolute Error (MAE) of 2.12, a Root Mean Square Error (RMSE) of 4.28, a Mean Absolute Percentage Error (MAPE) of 5.47%, and an $R^2$ score of 0.966. Compared with the strongest federated baseline, it improves MAE by up to 18.77%, RMSE by 16.41%, and MAPE by more than 22%, while reducing communication overhead through an 8:1 latent representation compression ratio. These results demonstrate the effectiveness of FedTraffic as a scalable, privacy-preserving, and interpretable solution for next-generation intelligent transportation systems. Full article

Primary prevention is essential to reduce disease burden before clinical onset, yet it remains less systematically integrated into care than diagnosis and treatment. Although artificial intelligence (AI) is increasingly used in medicine, most applications have focused on secondary and tertiary prevention, including diagnosis, prognostic stratification, and disease management, while its role in primary prevention remains less defined. This narrative review examines current AI applications across four modifiable lifestyle domains relevant to prevention and healthspan promotion: nutrition, physical activity, sleep, and mental health. We synthesize evidence on machine-learning models, wearable-derived algorithms, computer-vision tools, just-in-time adaptive interventions, and conversational agents used in consumer, community, and hybrid clinical–digital settings. AI applications support postprandial glycemic prediction, automated dietary assessment, meal-planning adherence, sedentary-pattern detection, personalized exercise recommendations, adaptive behavioral nudges, sleep monitoring, circadian-aware recommendations, psychoeducation, stress-management support, and early identification of psychological vulnerability. Collectively, these tools may extend prevention beyond episodic clinical encounters toward continuous, personalized, and context-aware support. However, evidence remains limited by short follow-up, reliance on surrogate or engagement outcomes, digitally literate populations, and insufficient validation in real-world preventive-care pathways. AI is therefore a promising enabling technology for proactive, healthspan-oriented medicine, provided future studies demonstrate long-term effectiveness, equity, safety, and responsible implementation. Full article

Differentiating intraocular tumors from inflammatory conditions remains a major diagnostic challenge in ophthalmology, particularly in masquerade syndromes where clinical and imaging features overlap. Despite advances in ocular imaging, no single modality provides sufficient diagnostic specificity. This narrative review synthesizes current evidence on optical imaging modalities, ultrasonography, and elastography, focusing on their complementary roles in intraocular disease characterization, with emphasis on shear wave elastography (SWE) within a multimodal framework. Conventional imaging provides partial insights: OCT and ultrasonography define structural features, while angiographic techniques assess vascular behavior. Elastography adds a biomechanical dimension by enabling in vivo assessment of tissue stiffness. Malignant lesions typically exhibit increased stiffness and heterogeneity, whereas inflammatory processes show more variable profiles, yet overlap exists. Integrating structural, vascular, and biomechanical data improves pattern recognition and diagnostic confidence. A tri-axis fusion model and a practical diagnostic workflow are proposed to support clinical decision-making. Multi-modality image fusion represents a shift toward hybrid multi-dimensional tissue characterization in ophthalmology. The addition of elastography may assist in the differential diagnosis process between intraocular tumors and inflammatory conditions. Further standardization, validation, and integration with artificial intelligence are required for routine clinical implementation. Full article

Addressing the critical challenge of signal degradation in biosensor cardiac monitoring, this paper introduces an efficient hybrid dilated–long short-term memory (LSTM) with residual learning (HDLR), which is a novel architecture engineered for a high-fidelity electrocardiogram (ECG) denoising signal. The proposed HDLR model synergistically integrates with dilated convolutions to expand the receptive field for multi-scale feature extraction. This is an LSTM-based backbone used to resolve long term temporal dependencies with residual learning paths to stabilize gradient flow and accelerate convergence. The proposed HDLR architecture integrates three core functional components with dilated convolutional layers utilized for local temporal feature extraction, where varying dilation rates expand the receptive field to capture both local waveform patterns and broader morphological structures without increasing computational complexity. Experimental results demonstrate a significant leap in performance, with the HDLR model achieving a mean squared error (MSE) of 0.002176, a signal-to-noise ratio (SNR) of 14.4420 dB, and a Matthews correlation coefficient (MCC) of 0.9822. Beyond quantitative metrics, the proposed HDLR architecture exhibits exceptional robustness in preserving cardiac morphology, specifically the P-wave and QRS complex of the ECG signal under stochastic noise conditions. These findings underscore the HDLR model’s potential as a backbone for next generation, real time diagnostic systems in intelligent healthcare. Full article

Radiotherapy is a cornerstone of breast cancer treatment, but its efficacy is frequently limited by intrinsic and acquired radioresistance as well as dose-limiting toxicity to surrounding normal tissues. Nanoparticle-mediated radiosensitization has emerged as a promising strategy to enhance the therapeutic index of irradiation by combining physical dose amplification with biological, microenvironmental, and immunological modulation. In this systematic review, we evaluated preclinical evidence on nanoparticle-mediated radiosensitization in breast cancer, with emphasis on nanoplatform design, mechanistic patterns, therapeutic efficacy, and translational relevance. A total of 66 studies published between 2015 and 2026 were included. The identified systems encompassed a broad range of materials, including gold-, silver-, platinum-, bismuth-, gadolinium-, polymer-, lipid-, and hybrid-based nanoplatforms, frequently incorporating targeting ligands, catalytic components, biomimetic coatings, or therapeutic payloads. Enhanced radiation responses were most commonly associated with high-atomic-number (high-Z)-mediated energy deposition, increased reactive oxygen species generation, and enhanced DNA damage persistence. Additional mechanisms, including redox modulation, hypoxia targeting, regulated cell death, and immune activation, reflect the evolution of nanoparticle-assisted radiotherapy from predominantly physical radioenhancement toward multifunctional physicobiological strategies. Triple-negative breast cancer models predominated throughout the literature. Across preclinical models, nanoparticle-assisted irradiation consistently improved clonogenic survival, tumor control, and, in selected studies, survival. However, substantial heterogeneity in study design and limited use of rigorous radiobiological endpoints restricted cross-study comparability. The available preclinical evidence indicates that the most promising nanoparticle-mediated radiosensitization strategies integrate physical dose enhancement with biologically active mechanisms targeting oxidative stress, hypoxia, persistent DNA damage, immune signaling, and tumor microenvironmental resistance. Collectively, these findings suggest that the field is evolving from predominantly physical radioenhancement toward multifunctional, mechanism-driven physicobiological strategies. However, clinical translation remains constrained by methodological heterogeneity and limited radiobiological validation, highlighting the need for standardized preclinical evaluation and clinically feasible nanoplatforms tailored to subtype-specific mechanisms of radioresistance. Full article

Urban traffic anomaly detection is essential for intelligent transportation systems, particularly in smart city environments where fast identification of abnormal events can improve road safety and traffic management. This work proposes a novel ULSTM-driven architecture that explicitly models temporal dependencies across consecutive traffic frames to achieve more stable and temporally coherent reconstructions. The proposed framework leverages sequential spatio-temporal representations to improve the distinction between normal traffic patterns and anomalous events. To further enhance reliability, we introduce a Hybrid Weighted Fusion strategy that synergistically combines structural, perceptual and pixel-wise metrics. The framework’s parameters are optimized using a Discrete Dirichlet Sampling approach, achieving a peak F1 Score of 70.28%. Evaluations were conducted on a manually curated traffic anomaly dataset with frame-level annotations. Experimental results demonstrate that the ULSTM framework significantly outperforms frame-independent generative models by suppressing high-frequency reconstruction noise, providing a robust solution for real-world smart city deployments. While highly effective in complex scenarios, the proposed framework is strictly applicable to highly dynamic traffic environments with active motion, as static background ensembles can degrade performance. Full article

High-precision mapping of tree species composition is essential for sustainable forest management, biodiversity assessment, and ecosystem monitoring. Airborne hyperspectral imagery provides rich spectral and spatial information that enables detailed species discrimination. However, traditional single-stream convolutional neural networks (CNNs) often fail to fully exploit multi-dimensional features and are susceptible to spectral redundancy. In this study, we propose the MS-CBAM-TSCNet (Multi-Stream Convolutional Block Attention Deep Neural Network), a novel architecture specifically designed for tree species classification using aerial hyperspectral data. The proposed model integrates three parallel processing streams: a 1D spectral branch for capturing reflectance signatures, a 2D spatial branch for modeling contextual patterns, and a 3D spectral–spatial branch enhanced with Convolutional Block Attention Modules (CBAMs) to adaptively recalibrate channel-wise and spatial–spectral features. An adaptive gated fusion mechanism with attention-based weighting is introduced to dynamically combine the complementary representations extracted from the three streams, improving robustness to spectral redundancy and class imbalance. The method was evaluated on an airborne HyMap hyperspectral dataset (125 bands, with 4 m spatial resolution) acquired over a mixed boreal forest in Karlsruhe, Germany, comprising five dominant tree species. Using five-fold cross-validation on an augmented dataset, the MS-CBAM-TSCNet achieved an overall accuracy of 96.8%, a Kappa coefficient of 0.96, and a macro F1-score of 0.966, outperforming conventional 1D, 2D, and 3D CNNs, as well as a hybrid CNN-SVM approach across all evaluation metrics. An ablation study further confirms the complementary contributions of the multi-stream architecture, CBAM attention, and adaptive gated fusion. Pixel-wise classification maps demonstrate improved boundary delineation and reduced misclassification in mixed stands, highlighting the effectiveness of the proposed framework for operational forest inventory and ecological monitoring. Full article