Search Results (14,434)

Understanding and predicting how tourists move through a city is a challenging task, as it involves a complex interplay of spatial, temporal, and social factors. Traditional recommender systems often rely on structured data, trying to capture the nature of the problem. However, recent advances in Large Language Models (LLMs) open new possibilities for reasoning over richer, text-based representations of user context, even without a dedicated pre-training phase. In this study, we investigate the potential of LLMs to interpret and predict tourist movements in a real-world application scenario involving tourist visits to Verona, a municipality in Northern Italy, between 2014 and 2023. We propose an incremental prompt engineering approach that gradually enriches the model input, from spatial features alone to richer behavioral information, including visit histories, time information, and user cluster patterns. The approach is evaluated using six open-source models, enabling us to compare their accuracy and efficiency across various levels of contextual enrichment. The results provide a first insight about the abilities of LLMs to incorporate spatio-temporal contextual factors, thus improving predictions, while maintaining computational efficiency. The analysis of the model-generated explanations completes the picture by adding an interpretability dimension that most existing next-PoI prediction solutions lack. Overall, the study demonstrates the potential of LLMs to integrate multiple contextual dimensions in tourism mobility, highlighting the possibility of a more text-oriented, adaptive, and explainable T-RS. Full article

In Transformer-based image restoration models, the self-attention mechanism often introduces attention noise from irrelevant contextual feature, hindering the recovery of underlying clear content. Although many methods have been proposed to suppress attention noise, we note that most existing approaches are often developed for general vision tasks and fail to generalize across remote sensing image dehazing, where large-scale spatial structures pose additional challenges for attention modeling. How to effectively model scale-aware attention to suppress redundant activations becomes crucial for remote sensing image dehazing. In this paper, we propose a scale-adaptive differential Transformer (SDTformer), an architecture designed to suppress attention noise through a differential attention mechanism, thereby improving reconstruction fidelity. Specifically, the model incorporates a scale-adaptive differential self-attention module, which models contextual dependencies across different spatial scales and reduces redundant contextual interference by computing differential attention maps. Additionally, a dynamic differential feed-forward network is proposed to adaptively select informative spatial features, strengthening feature aggregation. To further enhance feature representation, a gated fusion module is introduced to aggregate multi-scale features generated by different encoder blocks, which facilitates the learning process of each decoder block and improves the final reconstruction performance. Extensive experimental results on the commonly used benchmarks show that our method achieves favorable performance against state-of-the-art approaches. Full article

Ecosystem services (ESs) support human well-being, but their integrated assessment in urban green spaces remains challenging, particularly at the project scale, where finer spatial resolution (tens of meters) is required. Historical parks are complex socio-ecological systems with non-linear ES interactions. This study develops a design-oriented framework to assess how restoration interventions influence regulation, maintenance, and cultural ES potential provision. Indicators derived from field surveys and established models were selected according to CICES V5.2 and adapted to ecological and cultural features of historical parks. Survey units were defined for each ES section to enable a spatially explicit comparison between current and design scenarios. A normalized scoring system was applied to evaluate category-level changes and overall interaction patterns. The framework was tested on the restoration project of Monza Park (northern Italy). Results show a marked increase in cultural and regulation services (+28% and +17%, respectively), while maintenance services exhibited a slight decrease (−3%). These trends are reflected in the Cumulative Indicator Score (CIS), indicating an overall positive balance of ES provision in the design scenario. The Design Effectiveness Score (DES) showed consistently non-negative values (DES ≥ 0), reaching maximum effectiveness in transitions to woody vegetation (DES ≈ 1). The Synergy–Trade-off Score (STS) confirmed a general increase in ES supply across all categories, with a clear prevalence of synergies over trade-offs. The proposed framework supports the data-driven, spatially explicit evaluation of design alternatives and can guide decision-making in historical park restoration. Full article

Objective biomarkers for neurodevelopmental disorders remain an unmet clinical need. The electroretinogram (ERG), a non-invasive recording of the retinal response to light, has shown promise as a physiological marker for autism spectrum disorder (ASD) and attention deficit/hyperactivity disorder (ADHD), yet existing classification approaches based on time-domain and time–frequency features achieve limited accuracy in clinically relevant multi-group scenarios. This study introduces ERG-Graph, a novel graph signal processing (GSP) framework that transforms each ERG waveform into a weighted, undirected graph through amplitude quantization and temporal-adjacency connectivity. Nine topological and spectral features, including total load centrality, clique number, algebraic connectivity, and clustering coefficient, were extracted from each graph to characterize the structural dynamics of the signal. Using light-adapted ERG recordings from 278 participants (ASD = 77, ADHD = 43, ASD + ADHD = 21, Control = 137), we evaluated these features across binary, three-group, and four-group classification scenarios using seven machine learning classifiers with 10-fold subject-wise cross-validation. The proposed ERG-Graph features achieved balanced accuracies of 0.91 (ASD vs. control, males) and 0.88 (ADHD vs. control, females). Critically, fusing ERG-Graph with time-domain features yielded a balanced accuracy of 0.81 for three-group classification (ASD vs. ADHD vs. control), representing an 11-percentage-point improvement over the previous benchmark of 0.70. Statistical analysis confirmed significant topological differences between groups (Kruskal–Wallis, p < 0.001; Cliff’s delta: large effect sizes), and SHAP analysis revealed that graph-theoretic features dominated the top-ranked predictors. These results demonstrate that graph-based topological features capture discriminative information in the ERG waveform that is inaccessible to conventional signal analysis methods, advancing the development of objective biomarkers for neurodevelopmental disorder screening. Full article

Detecting drivable areas is a fundamental task in autonomous driving systems. Although semantic segmentation networks have demonstrated strong performance in segmenting drivable regions, two key challenges persist. First, acquiring sufficient contextual information in complex road scenarios remains difficult, often leading to segmentation errors. Second, the coarseness of extracted features may degrade accuracy even when texture information is available in RGB images. To address these issues, we propose an enhanced DeepLabv3+ algorithm called Split Convolution Selective Attention Network (SCSANet), which incorporates the Adaptive Kernel (AK) and Split Convolution Attention (SCA) modules. AK adaptively adjusts the receptive field to accommodate varying road scenarios, while SCA improves boundary clarity by enhancing channel interaction. In addition, we employ surface normals to provide complementary geometric information, thereby strengthening the ability of the network to recognize drivable areas. To compensate for the lack of publicly available datasets for closed or semi-closed scenarios, we introduce XMUROAD, a new dataset of binocular disparity images. Experiments on the XMUROAD dataset demonstrate that the proposed architectural improvements yield an mIoU gain of 1.63% under the same RGB input, and the full pipeline with surface normal input achieves improvements of 1.55% to 2.59% in mF1 and 2.94% to 4.83% in mIoU over state-of-the-art methods. Experiments on the KITTI dataset further verify the generalization capability of SCSANet, with improvements of 1.58% in mF1 and 2.88% in mIoU over state-of-the-art methods. The proposed method provides a practical approach for accurate drivable area detection in closed and semi-closed mobile-robot scenarios. Full article

Vigna umbellate, a typical edible and medicinal crop, is rich in polyphenolic compounds with antioxidant, antibacterial, anti-inflammatory, and lipid-regulating activities. However, traditional methods for polyphenol content detection rely on chemical analysis, which is cumbersome and time-consuming, making it difficult to meet the demands of high-throughput rapid detection. Although hyperspectral imaging technology offers the potential for non-destructive and rapid detection, existing analytical methods are often limited by issues such as high spectral band redundancy, insufficient feature extraction, and inadequate model stability, which constrain prediction accuracy and practical application potential. To address this, this study proposes a multi-scale residual convolutional neural network (MS-RCNN) based on competitive adaptive reweighted sampling (CARS) for feature band selection, combined with near-infrared hyperspectral imaging technology, to construct a rapid and non-destructive prediction model for the polyphenol content of Vigna umbellata. The model employs a parallel multi-scale convolutional module to extract spectral features with different receptive fields, and incorporates residual connections and adaptive pooling mechanisms to enhance feature reuse and robustness. Experiments compared the performance of partial least squares regression (PLSR), least squares support vector machine (LS-SVM), multi-scale convolutional neural network (MS-CNN), and MS-RCNN models. The results indicate that the MS-RCNN model based on CARS screening achieved the best prediction performance, with a coefficient of determination (R²) of 0.9467, a root mean square error of prediction (RMSEP) of 0.0448, and a residual predictive deviation (RPD) of 4.33. Compared with the optimal PLSR and LSSVM models, its R² values were improved by 0.2078 and 0.1119, respectively. In summary, the MS-RCNN model proposed in this study enables rapid, non-destructive, and accurate prediction of polyphenol content in Vigna umbellata, providing an efficient technical approach for quality detection of edible and medicinal crops. Full article

Motor neuron diseases (MNDs) encompass a clinically heterogeneous group of neurodegenerative conditions with varying impact on dexterity, mobility, decision making, respiratory and bulbar dysfunction. While consensus best-practice recommendations exist for genetic screening, diagnostic work-up, pharmacological and respiratory management, disease-specific facets of driving safety, assessment approaches and intervention strategies to support patients for safe driving have not been comprehensively reviewed. MNDs have unique, phenotype-specific clinical features, which are distinct form other neuromuscular conditions which necessitate a careful and systematic approach to evaluate driving safety. While MNDs are primarily associated with progressive motor impairment, extrapyramidal, cerebellar, cognitive, behavioural, and respiratory manifestations of the disease also affect driving safety and necessitate comprehensive driving assessments and individualised strategies to enable patients to continue to drive. The majority of existing papers focus on amyotrophic lateral sclerosis, and low-incidence MND phenotypes, such as PLS, SBMA, PPS, are glaringly understudied from a driving safety perspective despite the relatively slower progression of these conditions. Beyond the review of specific aspects of driving in MNDs, the main objective of this review paper is to raise awareness of non-motor aspects of MNDs with regard to driving safety and to explore viable strategies to support patients to maintain their independence. Despite the considerable differences in driving regulations around the globe, there are core, disease-specific aspects of MND which are universal. The careful consideration of these clinical factors, comprehensive domain-by-domain assessments, and the implementation of practical, individualised adaptations may enable patients to continue driving safely, maintain their independence and enhance their quality of life. Full article

Against the backdrop of smart cities and digital cultural tourism, the accurate prediction of urban passenger flow is of great significance for public security management and resource allocation. However, existing studies mostly rely on single data sources or only perform a simple concatenation of multi-source features, lacking systematic indicator system design. Meanwhile, weekly or monthly data are commonly used with coarse temporal granularity, making it difficult to capture short-term fluctuations and lag effects. To overcome these limitations, this paper collects the daily passenger flow data of Hangzhou from 15 March 2024 to 15 March 2025; integrates multi-dimensional factors such as keyword search trends across platforms, holidays and major events, and online public opinion; and constructs three daily characteristic indicators: online search index, humanistic–meteorological index, and textual sentiment index. The data denoising, dimensionality reduction, and sentiment quantification are realized through methods including SSA, PCA, and SnowNLP. On this basis, a hybrid CNN-GRU model integrated with the attention mechanism is proposed. An improved artificial bee colony (ABC) algorithm is adopted for global hyperparameter optimization, and a weighted hybrid loss function (JQHL) is introduced to enhance the model’s adaptability to extreme values. The results show that the ABC-CNN-GRU-Attention model, incorporating multi-dimensional indicators, outperforms traditional methods on evaluation metrics, including MAE, RMSE, MAPE, R², and RPD, demonstrating a higher prediction accuracy and robustness. Full article

With the explosive growth of the Internet of Things (IoT) and cloud-computing traffic, Intrusion Detection Systems (IDSs) have become a cornerstone of network security. However, modern traffic data often exhibits extreme class imbalance and long-tailed distributions, leading to persistently high miss rates for minority attack categories in Machine Learning (ML)-based IDSs. Conventional oversampling may introduce decision noise, whereas standard Generative Adversarial Networks (GANs) can suffer from training instability and mode collapse when modeling high-dimensional tabular traffic features. To address these challenges, we propose a high-fidelity traffic augmentation framework based on an Adaptive Hybrid-loss Conditional GAN (AH-CGAN). Specifically, AH-CGAN introduces an iteration-dependent adaptive gradient penalty (AGP) schedule to enforce the Lipschitz continuity constraint more effectively during training and incorporates a feature-matching objective to align intermediate critic representations between real and synthetic traffic. Experiments on the CIC-IDS2017 benchmark show that AH-CGAN generates distribution-consistent synthetic samples and that augmentation improves downstream detection across multiple classifiers. In particular, the weighted F1-score of Logistic Regression increases from 0.8237 to 0.8697 (Δ = +0.0460, i.e., +4.6%). Overall, the proposed approach enhances minority coverage in the feature space and can improve class separability, providing a practical solution for long-tailed IDS. Full article

Remote sensing visual question answering (RS-VQA) is essential to intelligent Earth observation, as it supports interactive querying of high-resolution aerial images. Many existing methods struggle with fine-detail geospatial reasoning with remote sensing (RS) scenes due to RS scenes having intrinsic multi-scale object variance and pronounced spatial heterogeneity. The models tend to rely more on the linguistic prior than reasoning based on visual evidence. In this paper, we present PMA-VQA, a progressive multi-scale feature fusion with spatially adaptive attention, to embed the RS-VQA task in spatially based hierarchical feature integration. For hierarchical, multi-level, language-informed integration, we propose a spatial attention aggregation module (SAAM) and a progressive feature fusion and classification module (PFCM). The SAAM employs spatially adaptive gating to align cross-modal features with semantic context, while the PFCM integrates multi-scale representations across high-level semantic abstractions and low-level space. The experimental results on RS-VQA LR and HR benchmarks validate that PMA-VQA outperformed all competing methods in terms of accuracy and robustness. Evaluation of HRVQA further confirmed the effectiveness of the SAAM and PFCM across diverse RS scenes. Full article

Quantitatively characterizing the dynamic evolution of fracture swarms under offset well low-frequency distributed acoustic sensing (LF-DAS) monitoring remains a significant challenge. This study proposes a physics-data dual-driven closed-loop inversion framework to address this problem. The framework consists of three core modules: (1) a fluid–solid coupled semi-analytical forward model applicable to variable-rate injection and shut-in conditions; (2) an automatic key feature identification method based on multi-scale scanning and physical polarity constraints; and (3) a dynamic inversion model for fracture swarms based on adaptive particle swarm optimization (APSO). Validation against the classical PKN model confirms that the proposed forward model accurately reproduces the fundamental fracture propagation behavior, with good agreement in fracture half-length and net pressure evolution. In synthetic inversion cases, the method successfully recovers the number of fractures, the dynamic flow rate allocation history, fracture length evolution, and the spatiotemporal strain rate response. A field application further demonstrates that three dominant fractures were generated during stimulation, reaching the vicinity of the monitoring well at 18, 27, and 46 min with corresponding spacings of approximately 21 m and 16 m. The proposed framework provides a new route for advancing LF-DAS monitoring from qualitative interpretation to quantitative dynamic inversion. Full article

Cross-domain scene classification aims to mitigate the distribution discrepancy between domains through domain adaptation techniques. With the rapid advancement of Vision–Language Models (VLMs), utilizing them for cross-domain scene classification has emerged as a promising research direction. Current methods utilize domain-specific prompts to facilitate domain adaptation through the CLIP model. However, for remote sensing images, the considerable differences in visual features across domains pose significant challenges for learning domain-specific prompts, leading to suboptimal cross-domain performance. In addition, they cannot reduce the domain shift that exists between the source domain and the target domain. To address the above challenges, we propose a novel cross-domain scene classification method, DIPVR (Domain-Invariant Prompts and Visual Representations), which enhances model performance by learning domain-invariant features for both prompts and visual representations. Specifically, we propose learning domain-invariant prompts and introducing prior knowledge to guide the prompt-learning process. To learn domain-invariant visual representations, we propose a Visual Invariant Learning module that adaptively extracts the shared features between the source and target domains. Finally, visual features are matched with context features to align the domain distributions between the source and target domains. The experimental results on the cross-domain scene classification datasets demonstrate that our proposed method outperforms the baseline methods, achieving optimal cross-domain transfer performance. Full article

Skeleton-based action recognition is a critical technology for intelligent sports analysis. Although the human skeletal structure exhibits inherent bilateral symmetry, sensor noise on resource-constrained edge devices frequently induces geometric distortion and topological asymmetry. Consequently, achieving a balance between high accuracy and real-time performance remains a significant challenge. To this end, we propose EMS-GCN, an Efficient Multi-scale Shift Graph Convolutional Network that integrates geometric priors. Specifically, we design a Gaussian kernel-driven topology refinement module to mitigate structural noise inherent in sensor data. By leveraging geometric symmetry and Gaussian distances among nodes, this module dynamically constrains graph topology learning, thereby effectively rectifying the structural asymmetry and ambiguity induced by noise. Furthermore, we construct a Multi-scale Shift Linear Attention (MSLA) module to replace computationally intensive temporal convolutions. Leveraging temporal shift invariance, this module captures multi-scale contexts via parameter-free shift operations. Furthermore, we introduce a linear temporal attention mechanism to model global temporal dependencies with linear complexity, effectively resolving the information asymmetry inherent in long-range interactions. Finally, EMS-GCN incorporates a dual-branch attention structure to adaptively calibrate feature responses. Extensive experiments demonstrate that our model maintains high recognition accuracy with only 0.56M parameters, representing a reduction of over 60% compared to mainstream baselines. These results validate the efficacy of leveraging geometric and temporal symmetries to enhance real-time sports analysis. Full article

This study develops a reduced-order predictive model of the Sit-To-Stand (STS) task to examine whether a simplified biomechanical representation can reproduce key STS patterns reported in the literature and to investigate the role played in movement by a flexible trunk. The model represents the human body as a planar multibody system and formulates STS as an optimization problem within a discrete mechanics framework. This formulation combines reduced model complexity, explicit torso flexibility, and a structure-preserving numerical approach for trajectory generation. Simulations were used to evaluate the effects of movement duration, reduced joint strength, and seat height on joint torques, kinematics, trunk motion, and ground reaction forces (GRFs). The results reproduced several qualitative trends reported in previous experimental studies, including increased peak joint torques and GRFs with shorter movement duration, lower joint strength, and reduced seat height, as well as greater compensatory trunk motion under more demanding conditions. These findings suggest that the proposed framework captures key adaptive features of STS mechanics and may provide useful insights for rehabilitation analysis and the design of assistive technologies such as lower-limb exoskeletons and rehabilitation devices. At the same time, the present work should be regarded as an initial methodological study, since validation is currently qualitative and further experimental calibration, quantitative validation, and sensitivity analysis remain part of ongoing work. Full article

Chronic stress is a time-dependent condition characterized by sustained dysregulation across neural, autonomic, and endocrine systems, with important consequences for both health and socioeconomic outcomes. Unlike acute stress, which is typically characterized by short-lived physiological activation, chronic stress reflects an accumulated allostatic load and a longer-term recalibration of stress response systems. Recent advances in physiological sensing and artificial intelligence (AI) have supported the development of computational approaches for chronic stress detection using electroencephalography (EEG), heart rate variability (HRV), photoplethysmography (PPG), electrodermal activity (EDA), and wearable multimodal platforms. This narrative review examines current AI-based studies through three main inferential paradigms: resting baseline dysregulation, longitudinal physiological monitoring, and reactivity-based inference. Across modalities, classical machine learning (ML) methods, particularly support vector machines (SVMs) and tree-based ensembles, remain the most commonly used approaches, largely because available datasets are small and most pipelines still depend on engineered features. Deep learning (DL) methods are beginning to emerge, but their use remains constrained by the lack of large, standardized, longitudinal datasets specifically designed for chronic stress research. Major challenges include ambiguity in stress labeling, limited longitudinal validation, circadian confounding, inter-individual variability, and small cohort sizes. Future progress will depend on standardized datasets, biologically grounded multimodal integration, hybrid baseline-reactivity modeling, adaptive personalization, and more interpretable AI systems. Greater emphasis is also needed on clinical relevance and generalizability if AI-based chronic stress monitoring is to move beyond experimental settings. Full article