Search Results (3,549)

Accurately predicting the next location of urban individuals is a central issue in human mobility research. Human mobility exhibits diverse patterns, requiring the integration of spatiotemporal contexts for location prediction. In this context, multi-view learning has become a prominent method in location prediction. Despite notable advances, current methods still face challenges in effectively capturing non-spatial proximity of regional preferences, complex temporal periodicity, and the ambiguity of location semantics. To address these challenges, we propose a representation-enhanced multi-view learning network (ReMVL-Net) for location prediction. Specifically, we propose a community-enhanced spatial representation that transcends geographic proximity to capture latent mobility patterns. In addition, we introduce a multi-granular enhanced temporal representation to model the multi-level periodicity of human mobility and design a rule-based semantic recognition method to enrich location semantics. We evaluate the proposed model using mobile phone data from Fuzhou. Experimental results show a 2.94% improvement in prediction accuracy over the best-performing baseline. Further analysis reveals that community space plays a key role in narrowing the candidate location set. Moreover, we observe that prediction difficulty is strongly influenced by individual travel behaviors, with more regular activity patterns being easier to predict. Full article

Neural networks (NNs) achieve breakthroughs in computer vision and natural language processing,yet their “black box” nature persists. Traditional methods prioritise parameter optimisation and loss design, overlooking NNs’ fundamental structure as topologically organised nonlinear computational systems. This work proposes a complex network theory framework decoding structure–function coupling by mapping convolutional layers, fully connected layers, and Dropout modules into graph representations. To overcome limitations of heuristic compression techniques, we develop a topology-sensitive adaptive pruning algorithm that evaluates critical paths via node strength centrality, preserving structural–functional integrity. On CIFAR-10, our method achieves 55.5% parameter reduction with only 7.8% accuracy degradation—significantly outperforming traditional approaches. Crucially, retrained pruned networks exceed original model accuracy by up to 2.63%, demonstrating that topology optimisation unlocks latent model potential. This research establishes a paradigm shift from empirical to topologically rationalised neural architecture design, providing theoretical foundations for deep learning optimisation dynamics. Full article

Bacterial communities within the female upper genital tract may influence the risk of ovarian cancer. In this retrospective cohort pilot study, we aim to detect different communities of bacteria between ovarian cancer and normal controls using topic modeling, a natural language processing tool. RNA was extracted and analyzed using the VITCOMIC2 pipeline. Topic modeling assessed differences in bacterial communities. Idatuning identified an optimal latent topic number and Latent Dirichlet Allocation (LDA) assessed topic differences between high-grade serous ovarian cancer (HGSOC) and controls. Results were validated using The Cancer Genome Atlas (TCGA) HGSOC dataset. A total of 801 unique taxa were identified, with 13 bacteria significantly differing between HGSOC and normal controls. LDA modeling revealed a latent topic associated with HGSOC samples, containing bacteria Escherichia/Shigella and Corynebacterineae. Pathway analysis using KEGG databases suggest differences in several biologic pathways including oocyte meiosis, aldosterone-regulated sodium reabsorption, gastric acid secretion, and long-term potentiation. These findings support the hypothesis that bacterial communities in the upper female genital tract may influence the development of HGSOC by altering the local environment, with potential functional implications between HGSOC and normal controls. However, further validation is required to confirms these associations and determine mechanistic relevance. Full article

This study focuses on analyzing progressive Type-II right censoring competing risks datasets. The latent causes of failures are assumed to follow independent generalized linear exponential distributions. The maximum likelihood and maximum product of spacing methods are employed to estimate the unknown parameters and survival indices. Furthermore, approximate confidence intervals are derived using the asymptotic normality of the maximum likelihood and the maximum product of spacing estimators. Additionally, bootstrap methods are employed to construct confidence intervals. A comprehensive simulation study is carried out to evaluate the effectiveness of these estimation approaches. Finally, real-world datasets are analyzed to illustrate the practical applicability of the proposed model. Full article

Soft data, such as seismic imagery, plays a critical role in subsurface modeling by providing indirect constraints away from hard data locations. However, validating whether subsurface model realizations honor this type of data remains a challenge due to the lack of robust quantitative tools. This study introduces a machine learning-based workflow for soft data checking that uses an autoencoder (AE) to encode 2D seismic slices into a latent space. Subsurface model realizations are transformed into the same domain and projected into this latent space, enabling both visual and quantitative comparisons using principal component analysis and Euclidean distances. We demonstrate the workflow on rule-based models and their associated synthetic seismic data (soft data), showing that models with similar Markov chain parameters to the reference soft data score higher in proximity metrics. This approach provides a scalable, quantitative, and interpretable framework for evaluating the consistency between soft data and subsurface models, supporting better decision-making in reservoir characterization and other geoscience applications. Full article

Sustainable education requires addressing the challenges posed by digital transformation, including technostress among university students. This study evaluates technostress levels in higher education through the validation of the TS4US scale and its implications for sustainable learning environments. A cross-sectional study was conducted with 328 university students from four districts in Lima, Peru, using an online survey to measure technostress. Confirmatory factor analysis (CFA) was performed to assess the psychometric properties of the TS4US scale, resulting in a refined model with two latent factors and thirteen validated items. Findings indicate that 28% of students experience high technostress levels, while 5% report very high levels, though no significant associations were found between technostress and sociodemographic variables such as campus location, employment status, gender, and academic level. The TS4US instrument had been previously validated in Chile; this study confirms its structure in a new sociocultural context, reinforcing its cross-cultural applicability. These results highlight the need for sustainable strategies to mitigate technostress in higher education, including institutional support, digital literacy programs, and policies fostering a balanced technological environment. Addressing technostress is essential for promoting sustainable education (SDG4) and enhancing student well-being (SDG3). This study directly contributes to the achievement of Sustainable Development Goals 3 (Good Health and Well-being) and 4 (Quality Education) by providing validated tools and evidence-based recommendations to promote mental health and equitable access to digital education in Latin America. Future research should explore cross-country comparisons and targeted interventions, including digital well-being initiatives and adaptive learning strategies, to ensure a resilient and sustainable academic ecosystem. Full article

This paper examines the farming systems and operational structures in the Region of Western Macedonia (RWM), Greece and constructs a typology of farms based on structural, operational, and socio-economic characteristics. Agriculture remains a vital pillar of the regional economy, particularly in the context of RWM’s ongoing transition to a post-lignite development model. Using farm-level data from the 2018 Farm Accountancy Data Network (FADN), Principal Component Analysis (PCA) identified four latent dimensions of farm heterogeneity—income and productivity, asset base, land size, and labour structure. Hierarchical and K-means cluster analysis revealed three distinct farm types: (1) medium-sized, high-efficiency farms with moderate reliance on subsidies (30% of the sample); (2) small-scale, family farms with modest productivity and limited capitalisation (48%); and (3) large, asset-rich farms exhibiting structural inefficiencies and lower output per hectare (22%). These findings highlight structural vulnerabilities, particularly the predominance of undercapitalised smallholdings, and provide a data-driven foundation for Thdesigning differentiated policies that support farm resilience, generational renewal, and sustainable rural development. Full article

As an integral component of humanity’s cultural heritage, traditional villages universally confront challenges such as population loss and cultural discontinuity amid rapid urbanization. Cluster-based protection models have increasingly become the international consensus for addressing the survival crisis of such settlements. This study selects the Central Yunnan region of Southwest China—characterized by its complex geography and multi-ethnic habitation—as the research area. Employing ArcGIS spatial analysis techniques alongside clustering algorithms, we examine the spatial distribution characteristics and clustering patterns of 251 traditional villages within this region. The findings are as follows. In terms of spatial distribution, traditional villages in Central Yunnan are unevenly dispersed, predominantly aggregating on mid-elevation gentle slopes; their locations are chiefly influenced by rivers and historical courier routes, albeit with only indirect dependence on waterways. Regarding single-cluster attributes, the spatial and geomorphological features exhibit a composite “band-and-group” pattern shaped by river valleys; culturally, two dominant modes emerge—“ancient-route-dependent” and “ethnic-symbiosis”—reflecting an economy-driven cultural mechanism alongside latent marginalization risks. Concerning construction characteristics, the “Qionglong-Ganlan” and Han-style “One-seal” residential features stand out, illustrating both adaptation to mountainous environments and the cumulative effects of historical culture. Based on these insights, we propose a three-tiered clustering classification framework—“comprehensive-element coordination”, “feature-led”, and “potential-cultivation”—to inform the development of contiguous and typological protection strategies for traditional villages in highland, multi-ethnic regions. Full article

This study presents a topology optimization approach to enhance the discharging performance of a latent heat thermal energy storage (LHTES) system using paraffin wax as the phase-change material (PCM) and a high-conductivity aluminium structure. Solidification is primarily governed by conduction, and the average heat transfer rate during this process is significantly lower than during melting; therefore, the optimization focused on the discharge phase. In a previous study, a novel LHTES device based on a Cartesian lattice was investigated experimentally and numerically. The validated numerical model from that study was adopted as the reference and used in a 2D topology optimization study based on the Solid Isotropic Material with Penalization (SIMP) method. The objective was to promote more uniform temperature distribution and reduce discharging time while maintaining the same aluminium volume fraction as in the reference device. Topology optimization produced a branched fin design, which was then extruded into a 3D model for comparison with the reference geometry. The optimized design resulted in improved temperature uniformity and a faster solidification process. Specifically, the time required to solidify 90% of the PCM was reduced by 12.3%, while the time to release 90% of the latent heat during the solidification process improved by 7.6%. Full article

The role of emotional intelligence (EI) in the academic context has been steadily established, together with its impact on students’ academic achievement, well-being, and professional success. Therefore, this study examined the development of a key EI ability—emotional understanding—throughout secondary school and explored its impact on students’ academic achievement (maternal language and mathematics) at the end of this cycle, using the Vocabulary of Emotions Test. A total of 222 students were followed over the entire 3-year secondary cycle, using a three-wave longitudinal design spanning from 10th to 12th grade. At the first wave, participants were aged between 14 and 18 years (M = 15.4; SD = 0.63), with 58.6% being female. Overall, the results of Latent Growth Curve modeling indicated that students’ emotional understanding increased over the secondary school cycle. While student’s gender predicted the emotional understanding change patterns throughout secondary school, student’s GPA in 10th grade did not. Moreover, the initial levels of ability-based emotional understanding predicted students’ achievement in maternal language at the end of the cycle. Our findings offer valuable insights into how EI skills can contribute to academic endeavors in late adolescence and will explore their impact on educational settings. Full article

Composite laminates are widely used in aerospace, automotive, construction, and luxury industries, owing to their superior mechanical properties and design flexibility. However, detecting manufacturing defects and in-service damage remains a vital challenge for structural safety. While traditional unsupervised machine learning methods have been used in structural health monitoring (SHM), their high false positive rates limit their reliability in real-world applications. This issue is mostly inherited from their limited ability to capture small temporal variations in Lamb wave signals and their dependence on shallow architectures that suffer with complex signal distributions, causing the misclassification of damaged signals as healthy data. To address this, we suggested an unsupervised anomaly detection framework that integrates a self-attention autoencoder with a Gaussian mixture model (SAE-GMM). The model is solely trained on healthy Lamb wave signals, including high-quality synthetic data generated via a generative adversarial network (GAN). Damages are detected through reconstruction errors and probabilistic clustering in the latent space. The self-attention mechanism enhances feature representation by capturing subtle temporal dependencies, while the GMM enables a solid separation among signals. Experimental results demonstrated that the proposed model (SAE-GMM) achieves high detection accuracy, a low false positive rate, and strong generalization under varying noise conditions, outperforming traditional and deep learning baselines. Full article

End-to-end relation extraction (E2ERE) generally performs named entity recognition and relation extraction either simultaneously or sequentially. While numerous studies on E2ERE have centered on enhancing span representations to improve model performance, challenges remain due to the gaps between subtasks (named entity recognition and relation extraction) and the modeling discrepancies between entities and relations. In this paper, we propose a novel Label Annotation Interaction-based representation enhancement method for E2ERE, which institutes a two-phase semantic interaction to augment representations. Specifically, we firstly feed label annotations that are easy to manually annotate into a language model, and conduct the first-round interaction between three types of tokens with a partial attention mechanism; Then we construct a latent multi-view graph to capture various possible links between label and entity (pair) nodes, facilitating the second-round interaction between entities and labels. A series of comparative experiments with methods of various transformer-based architectures currently in use show that LAI-Net can maintain performance on par with the current SOTA in terms of NER task, and achieves significant improvements over existing SOTA models in terms of RE task. Full article

Over recent decades, phase-change materials (PCMs) have gained prominence as latent-heat thermal energy storage systems in building envelopes because of their high energy density. However, only PCMs that complete a full daily charge–discharge cycle can deliver meaningful energy and carbon-emission savings. This simulation study introduces a methodology that simultaneously optimizes PCM integration for storage efficiency, indoor thermal comfort, and energy savings. Two new indicators are proposed: overall storage efficiency (EC_n), which consolidates heating and cooling-efficiency ratios into a single value, and the performance factor (PF), which quantifies the PCM’s effectiveness in maintaining thermal comfort. Using EnergyPlus v8.9 coupled with DesignBuilder, a residential ASHRAE 90.1 mid-rise apartment was modeled in six warm-temperate (Cfb) European cities for the summer period from June 1 to August 31. Four paraffin PCMs (RT-22/25/28/31 HC, 20 mm thickness) were tested under natural and controlled ventilation strategies, with windows opening 50% when outdoor air was at least 2 °C cooler than indoors. Simulation outputs were validated against experimental cubicle data, yielding a mean absolute indoor temperature error ≤ 4.5%, well within the ±5% tolerance commonly accepted for building thermal simulations. The optimum configuration—RT-25 HC with temperature-controlled ventilation—achieved PF = 1.0 (100% comfort compliance) in all six cities and delivered summer cooling-energy savings of up to 3376 kWh in Paris, the highest among the locations studied. Carbon-emission reductions reached 2254 kg CO₂-e year⁻¹, and static payback periods remained below the assumed 50-year building life at a per kg PCM cost of USD 1. The EC_n–PF framework, therefore, provides a transparent basis for selecting cost-effective, energy-efficient, and low-carbon PCM solutions in warm-temperate buildings. Full article

High-rise construction presents heightened safety risks due to vertical complexity, spatial constraints, and workforce variability. Conventional safety management often proves insufficient, especially in rapidly urbanizing or resource-limited settings. This study proposes a hybrid methodological framework to systematically identify and prioritize the critical factors influencing occupational safety in Vietnamese high-rise construction projects. Based on 181 valid survey responses from construction professionals, 23 observed variables were developed through extensive literature review and expert consultation. Exploratory Factor Analysis (EFA) was employed to empirically group 23 validated indicators into five key latent dimensions: (1) Safety Training and Inspection, (2) Employer’s Knowledge and Responsibility, (3) Worker’s Competence and Compliance, (4) Working Conditions and Environment, and (5) Safety Equipment and Signage. These dimensions were then structured into an Analytic Hierarchy Process (AHP) model, with pairwise comparisons conducted by industry experts to calculate consistency ratios and derive factor weights across three high-rise project case studies. The findings provide actionable insights for construction managers, safety professionals, and policymakers in developing and underdeveloped countries, supporting data-driven decision-making for safer and more sustainable urban development. Full article

Bruising in ‘Korla’ pears represents a prevalent phenomenon that leads to progressive fruit decay and substantial economic losses. The detection of early-stage bruising proves challenging due to the absence of visible external characteristics, and existing deep learning models have limitations in weak feature extraction under complex optical interference. To address the postharvest latent damage detection challenges in ‘Korla’ pears, this study proposes a collaborative detection framework integrating structured-illumination reflectance imaging (SIRI) with multi-order gated attention mechanisms. Initially, an SIRI optical system was constructed, employing 150 cycles·m⁻¹ spatial frequency modulation and a three-phase demodulation algorithm to extract subtle interference signal variations, thereby generating RT (Relative Transmission) images with significantly enhanced contrast in subsurface damage regions. To improve the detection accuracy of latent damage areas, the MOGA-UNet model was developed with three key innovations: 1. Integrate the lightweight VGG16 encoder structure into the feature extraction network to improve computational efficiency while retaining details. 2. Add a multi-order gated aggregation module at the end of the encoder to realize the fusion of features at different scales through a special convolution method. 3. Embed the channel attention mechanism in the decoding stage to dynamically enhance the weight of feature channels related to damage. Experimental results demonstrate that the proposed model achieves 94.38% mean Intersection over Union (mIoU) and 97.02% Dice coefficient on RT images, outperforming the baseline UNet model by 2.80% with superior segmentation accuracy and boundary localization capabilities compared with mainstream models. This approach provides an efficient and reliable technical solution for intelligent postharvest agricultural product sorting. Full article