Search Results (13,985)

Knowledge-graph-based recommendation aims to provide personalized recommendation services to users based on their historical interaction information, which is of great significance for shopping transaction rates and other aspects. With the rapid growth of online shopping, the knowledge graph constructed from users’ historical interaction data now incorporates multiattribute information, including timestamps, images, and textual content. The information of multiple modalities is difficult to effectively utilize due to their different representation structures and spaces. The existing methods attempt to utilize the above information through simple embedding representation and aggregation, but ignore targeted representation learning for information with different attributes and learning effective weights for aggregation. In addition, existing methods are not sufficient for effectively modeling temporal information. In this article, we propose MTR, a knowledge graph recommendation framework based on mixture of experts network. To achieve this goal, we use a mixture-of-experts network to learn targeted representations and weights of different product attributes for effective modeling and utilization. In addition, we effectively model the temporal information during the user shopping process. A thorough experimental study on popular benchmarks validates that MTR can achieve competitive results. Full article

Resource-exhausted cities have long played a key role in national energy development. Urban renewal projects, such as the renovation of old underground commercial spaces, can improve urban vitality and promote sustainable development. However, in resource-based cities, traditional industries dominate, while new industries such as modern commerce develop slowly. This results in low economic dynamism and weak motivation for urban development. To address this issue, we propose a systematic method for analyzing construction risks during the decision-making stage of renovation projects. The method includes three steps: risk value assessment, risk factor identification, and risk weight calculation. First, unlike previous studies that only used SWOT for risk factor analysis, we also applied it for project value assessment. Then, using the Work Breakdown Structure–Risk Breakdown Structure framework method (WBS-RBS), we identified specific risk sources by analyzing key construction technologies throughout the entire lifecycle of the renovation project. Finally, to enhance expert consensus, we proposed an improved Delphi–Analytic Hierarchy Process method (Delphi–AHP) to calculate risk indicator weights for different construction phases. The risk analysis covered all lifecycle stages of the renovation and upgrading project. The results show that in the Fushun city renovation case study, the established framework—consisting of five first-level indicators and twenty s-level indicators—enables analysis of renovation projects. Among these, management factors and human factors were identified as the most critical, with weights of 0.3608 and 0.2017, respectively. The proposed method provides a structured approach to evaluating renovation risks, taking into account the specific characteristics of construction work. This can serve as a useful reference for ensuring safe and efficient implementation of underground commercial space renovation projects in resource-exhausted cities. Full article

Landslides are destructive geological hazards, making accurate landslide susceptibility assessment essential for disaster prevention and mitigation. However, existing studies often lack scientific rigor in negative sample construction and have unclear model applicability. This study focuses on Ya’an City, Sichuan Province, China, and proposes an innovative approach to negative sample construction using Geographically Weighted Regression (GWR), which is then integrated with Tabular Network (TabNet), a deep learning architecture tailored to structured tabular data, to assess landslide susceptibility. The performance of TabNet is compared against Random Forest, Light Gradient Boosting Machine, deep neural networks, and Residual Networks. The experimental results indicate that (1) the GWR-based sampling strategy substantially improves model performance across all tested models; (2) TabNet trained using the GWR-based negative samples achieves superior performance over all other evaluated models, with an average AUC of 0.9828, exhibiting both high accuracy and interpretability; and (3) elevation, land cover, and annual Normalized Difference Vegetation Index are identified as dominant predictors through TabNet’s feature importance analysis. The results demonstrate that combining GWR and TabNet substantially enhances landslide susceptibility modeling by improving both accuracy and interpretability, establishing a more scientifically grounded approach to negative sample construction, and providing an interpretable, high-performing modeling framework for geological hazard risk assessment. Full article

Manganese (Mn) excess and low pH often coexist in some citrus orchard soils. Little information is known about the underlying mechanism by which raising pH reduces Mn toxicity in citrus plants. ‘Sour pummelo’ (Citrus grandis (L.) Osbeck) seedlings were treated with 2 (Mn2) or 500 (Mn500) μM Mn at a pH of 3 (P3) or 5 (P5) for 25 weeks. Raising pH mitigated Mn500-induced increases in Mn, iron, copper, and zinc concentrations in roots, stems, and leaves, as well as nitrogen, phosphorus, potassium, calcium, magnesium, sulfur, copper, iron, and zinc distributions in roots, but it mitigated Mn500-induced decreases in nitrogen, phosphorus, potassium, calcium, magnesium, sulfur, and boron concentrations in roots, stems, and leaves, as well as nutrient imbalance. Raising pH mitigated Mn500-induced necrotic spots on old leaves, yellowing of young leaves, decreases in seedling growth, leaf chlorophyll concentration, and CO₂ assimilation (A_CO2), increase in root dry weight (DW)/shoot DW, and alterations of leaf chlorophyll a fluorescence (OJIP) transients and related indexes. Further analysis indicated that raising pH ameliorated Mn500-induced impairment of nutrient homeostasis, leaf thylakoid structure by iron deficiency and competition of Mn with magnesium, and photosynthetic electron transport chain (PETC), thereby reducing Mn500-induced declines in A_CO2 and subsequent seedling growth. These results validated the hypothesis that raising pH reduced Mn toxicity in ‘Sour pummelo’ seedlings by (a) reducing Mn uptake, (b) efficient maintenance of nutrient homeostasis under Mn stress, (c) reducing Mn excess-induced impairment of thylakoid structure and PEPC and inhibition of chlorophyll biosynthesis, and (d) increasing A_CO2 and subsequent seedling growth under Mn excess. Full article

Understanding vehicle emissions is essential for developing effective carbon reduction strategies in the transport sector. Conventional emission models often assume homogeneity and linearity, overlooking real-world asymmetries that arise from variations in vehicle design and powertrain configurations. This study explores how machine learning and explainable AI techniques can effectively capture both symmetric and asymmetric emission patterns across different vehicle types, thereby contributing to more sustainable transport planning. Addressing a key gap in the existing literature, the study poses the following question: how do structural and behavioral factors contribute to asymmetric emission responses in internal combustion engine vehicles compared to new energy vehicles? Utilizing a large-scale Spanish vehicle registration dataset, the analysis classifies vehicles by powertrain type and applies five supervised learning algorithms to predict CO₂ emissions. SHapley Additive exPlanations (SHAPs) are employed to identify nonlinear and threshold-based relationships between emissions and vehicle characteristics such as fuel consumption, weight, and height. Among the models tested, the Random Forest algorithm achieves the highest predictive accuracy. The findings reveal critical asymmetries in emission behavior, particularly among hybrid vehicles, which challenge the assumption of uniform policy applicability. This study provides both methodological innovation and practical insights for symmetry-aware emission modeling, offering support for more targeted eco-design and policy decisions that align with long-term sustainability goals. Full article

Infection control and bleeding management in deep wounds remain urgent and unmet clinical challenges that demand innovative, multifunctional, and sustainable solutions. Unlike previously reported sodium alginate and silk fibroin-based gel formulations, the present work introduces a dual-functional system combining antimicrobial and haemostatic activity in the form of conformable aerogel beads. This dual-functional formulation is designed to absorb exudate, promote clotting, and provide localized antimicrobial action, all essential for accelerating wound repair in high-risk scenarios within a single biocompatible system. Aerogel beads were obtained by supercritical drying of a silk fibroin–sodium alginate blend, resulting in highly porous, spherical structures measuring 3–4 mm in diameter. The formulations demonstrated efficient ciprofloxacin encapsulation (42.75–49.05%) and sustained drug release for up to 12 h. Fluid absorption reached up to four times their weight in simulated wound fluid and was accompanied by significantly enhanced blood clotting, outperforming a commercial haemostatic dressing. These findings highlight the potential of silk-based aerogel beads as a multifunctional wound healing platform that combines localized antimicrobial delivery, efficient fluid and exudate management, biodegradability, and superior haemostatic performance in a single formulation. This work also shows for the first time how the prilling encapsulation technique with supercritical drying is able to successfully produce silk fibroin and sodium alginate composite aerogel beads. Full article

The accelerating integration of electric vehicles (EVs) into contemporary transportation infrastructure has underscored significant limitations in traditional charging paradigms, particularly in accommodating heterogeneous user requirements within dynamic operational environments. This study presents a differentiated optimization framework for EV charging strategies through the systematic classification of user types. A multidimensional decision-making environment is established for three representative user categories—residential, commercial, and industrial—by synthesizing time-variant electricity pricing models with dynamic carbon emission pricing mechanisms. A bi-level optimization architecture is subsequently formulated, leveraging deep reinforcement learning (DRL) to capture user-specific demand characteristics through customized reward functions and adaptive constraint structures. Validation is conducted within a high-fidelity simulation environment featuring 90 autonomous EV charging agents operating in a metropolitan parking facility. Empirical results indicate that the proposed typology-driven approach yields a 32.6% average cost reduction across user groups relative to baseline charging protocols, with statistically significant improvements in expenditure optimization (p < 0.01). Further interpretability analysis employing gradient-weighted class activation mapping (Grad-CAM) demonstrates that the model’s attention mechanisms are well aligned with theoretically anticipated demand prioritization patterns across the distinct user types, thereby confirming the decision-theoretic soundness of the framework. Full article

Background: Comorbid personality disorders (PDs) in patients with anorexia nervosa (AN) are associated with increased psychopathology, higher suicide risk, and poorer treatment response and outcomes. This study aimed to examine associations between gray matter (GM) volume and PDs in female adolescents with AN before and after short-term psychotherapeutic and nutritional therapy. Methods: Eighteen female adolescents with acute AN, mean age 15.9 years, underwent 3T magnetic resonance imaging before and after weight restoration. The average interval between scans was 2.6 months. Structural brain changes were analyzed using voxel-based morphometry. PDs were assessed using the Structured Clinical Interview for DSM-IV Axis II Disorders (SCID II) and the Assessment of Identity Development Questionnaire. Results: SCID-II total scores showed significant positive associations with GM volume in the mid-cingulate cortex at both time points and in the left superior parietal–occipital lobule at baseline. The histrionic subscale correlated with GM volume in the thalamus bilaterally and the left superior parietal–occipital lobule in both assessments, as well as with the mid-cingulate cortex at follow-up. Borderline and antisocial subscales were associated with GM volume in the thalamus bilaterally at baseline and in the right mid-cingulate cortex at follow-up. Conclusions: PDs in female adolescent patients with AN may be specifically related to GM alterations in the thalamus, cingulate, and parieto-occipital regions, which are present during acute illness and persist after weight restoration therapy. Full article

The abandoned goaf resulting from coal resource integration in China poses a significant threat to coal mine safety. The transient electromagnetic method (TEM) has emerged as a crucial technology for detecting goafs in coal mines due to its adaptable equipment and efficient implementation. In recent years, small-loop TEM has demonstrated high resolution and adaptability in challenging terrains with vegetation, such as coal mine ponding areas, karst regions, and reservoir seepage scenarios. By considering the sedimentary characteristics of coal seams and addressing the resistivity changes encountered in single-point inversion, a joint optimization inversion process incorporating lateral weighting factors and vertical roughness constraints has been developed to enhance the connectivity between adjacent survey points and improve the continuity of inversion outcomes. Through an OCCAM inversion approach, the regularization factor is dynamically determined by evaluating the norms of the data objective function and model objective function in each iteration, thereby reducing the reliance of inversion results on the initial model. Using the Xiaolong Coal Mine as a geological context, the impact of lateral and vertical weighting factors on the inversion outcomes of high- and low-resistivity structural models is examined through a control variable method. The analysis reveals that optimal inversion results are achieved with a combination of a lateral weighting factor of 0.5 and a vertical weighting factor of 0.1, ensuring both result continuity and accurate depiction of vertical and lateral electrical interfaces. The practical application of this approach validates its effectiveness, offering theoretical support and technical assurance for old goaf detection in coal mines, thereby holding significant engineering value. Full article

Effective earthquake early warning (EEW) is essential for disaster prevention in the built environment, enabling a rapid structural response, system shutdown, and occupant evacuation to mitigate damage and casualties. However, most current EEW systems lack rigorous reliability analyses of their predictive outcomes, limiting their effectiveness in real-world scenarios—especially for on-site warnings, where data are limited and time is critical. To address these challenges, we propose a Bayesian neural network (BNN) framework based on Stein variational gradient descent (SVGD). By performing Bayesian inference, we estimate the posterior distribution of the parameters, thus outputting a reliability analysis of the prediction results. In addition, we incorporate a continual learning mechanism based on elastic weight consolidation, allowing the system to adapt quickly without full retraining. Our experiments demonstrate that our SVGD-BNN model significantly outperforms traditional peak displacement (Pd)-based approaches. In a 3 s time window, the Pearson correlation coefficient R increases by 9.2% and the residual standard deviation $SD$ decreases by 24.4% compared to a variational inference (VI)-based BNN. Furthermore, the prediction variance generated by the model can effectively reflect the uncertainty of the prediction results. The continual learning strategy reduces the training time by 133–194 s, enhancing the system’s responsiveness. These features make the proposed framework a promising tool for real-time, reliable, and adaptive EEW—supporting disaster-resilient building design and operation. Full article

Synthetic aperture radar (SAR), as an active microwave imaging system, has the capability of all-weather and all-time observation. In response to the challenges of aircraft detection in SAR images due to the complex background interference caused by the continuous scattering of airport buildings and the demand for real-time processing, this paper proposes a YOLOv7-MTI recognition model that combines the attention mechanism and involution. By integrating the MTCN module and involution, performance is enhanced. The Multi-TASP-Conv network (MTCN) module aims to effectively extract low-level semantic and spatial information using a shared lightweight attention gate structure to achieve cross-dimensional interaction between “channels and space” with very few parameters, capturing the dependencies among multiple dimensions and improving feature representation ability. Involution helps the model adaptively adjust the weights of spatial positions through dynamic parameterized convolution kernels, strengthening the discrete strong scattering points specific to aircraft and suppressing the continuous scattering of the background, thereby alleviating the interference of complex backgrounds. Experiments on the SAR-AIRcraft-1.0 dataset, which includes seven categories such as A220, A320/321, A330, ARJ21, Boeing737, Boeing787, and others, show that the mAP and mRecall of YOLOv7-MTI reach 93.51% and 96.45%, respectively, outperforming Faster R-CNN, SSD, YOLOv5, YOLOv7, and YOLOv8. Compared with the basic YOLOv7, mAP is improved by 1.47%, mRecall by 1.64%, and FPS by 8.27%, achieving an effective balance between accuracy and speed, providing research ideas for SAR aircraft recognition. Full article

Intertrochanteric fractures in the elderly present complex challenges due to poor bone quality and comorbidities. Cephalomedullary (CM) nails offer biomechanical advantages that support early mobilization, yet complications such as cutout, implant failure, and malalignment persist. This review examines the effectiveness of CM nail fixation in geriatric extracapsular hip fractures and introduces the RESCUE technique—a structured, mnemonic-based approach aimed at improving surgical outcomes and reducing common complications. RESCUE stands for Reduce, Entry point, Screw, Compress, Unleash traction, and Enhance full-weight bearing. This six-step framework addresses the critical elements of fixation, including precise reduction, optimal entry point selection, central screw placement, controlled fracture compression, cautious traction management, and early mobilization. Case illustrations of frequent failure patterns underscore the practical application of the RESCUE technique. By following this systematic approach, surgeons can enhance construct stability, minimize failure risk, and promote functional recovery in elderly patients. Full article

Objective: To determine reference values of liver stiffness during the first week of extrauterine life in healthy newborns, according to gestational age, sex, and birth weight, using three elastography techniques: point shear wave elastography (pSWE) and two-dimensional shear wave elastography (2D-SWE) with convex and linear probes. Materials and Methods: This was a cross-sectional observational study conducted at a single center on a hospital-based cohort of 287 newborns between 24 and 42 weeks of gestation, admitted between January 2023 and May 2024. Cases with liver disease, significant neonatal morbidity, or technically invalid studies were excluded. Hepatic elastography was performed during the first week of life using pSWE and 2D-SWE with both convex and linear probes. Clinical and technical neonatal variables were recorded. Liver stiffness values were analyzed in relation to gestational age, birth weight, and sex. Linear regression models were applied to assess associations, considering p-values < 0.05 as statistically significant. Results: After applying exclusion criteria, valid liver stiffness measurements were obtained in 208 cases with pSWE, 224 with 2D-SWE (convex probe), and 222 with 2D-SWE (linear probe). A statistically significant inverse association between liver stiffness and gestational age (p < 0.03) was observed across all techniques except for 2D-SWE with the linear probe. Only 2D-SWE with the convex probe showed a significant association with birth weight. No significant differences were observed based on neonatal sex. The 2D-SWE technique with the convex probe demonstrated significantly shorter examination times compared to pSWE (p < 0.001). Conclusions: Neonatal liver stiffness measured by pSWE and 2D-SWE with a convex probe shows an inverse correlation with gestational age, potentially reflecting the structural and functional maturation of the liver. These techniques are safe, reliable, and provide useful information for distinguishing normal findings in preterm neonates from early hepatic pathology. The values obtained represent a valuable reference for clinical hepatic assessment in the neonatal period. Full article

Background: Previous research has demonstrated that 20 kHz probe or 37 kHz bath sonication of poloxamers comprising polypropylene glycol (PPG) and polyethylene glycol (PEG) blocks can generate degradation byproducts that are toxic to mammalian cells and organisms. Herein, an investigation of a PEGylated phospholipid micelle was undertaken to identify low-molecular-weight sonolytic degradation byproducts that could be cytotoxic. The concern here lies with the fact that sonication is a frequently employed step in drug delivery manufacturing processes, during which PEGylated phospholipids can be subjected to shear forces and other extreme oxidative and thermal conditions. Methods: Control and 20 kHz-sonicated micelles of DSPE-mPEG2000 were analyzed using dynamic light scattering (DLS) and zeta potential analyses to study colloidal properties, matrix-assisted laser desorption/ionization–time of flight (MALDI-TOF) mass spectroscopy (MS) and proton nuclear magnetic resonance (¹H-NMR) spectroscopy to study the structural integrity of DSPE-mPEG2000, and ¹H-NMR spectroscopy and high-performance liquid chromatography (HPLC) with ultraviolet (UV) detection to quantitate the formation of low-molecular-weight degradation byproducts. Results: MALDI-TOF-MS analyses of 20 kHz-sonicated DSPE-mPEG2000 revealed the loss of ethylene glycol moieties in accordance with depolymerization of the PEG chain; ¹H-NMR spectroscopy showed the presence of formate, a known oxidative/thermal degradation product of PEG; and HPLC-UV showed that the generation of formate was dependent on 20 kHz probe sonication time between 5 and 60 min. Conclusions: It was found that 20 kHz sonication can degrade the PEG chain of DSPE-mPEG2000, altering the micelle’s PEG corona and generating formate, a known ocular toxicant. Full article

E-commerce growth has reshaped consumer behavior and retail services, driving parcel demand and challenging last-mile logistics. Existing research predominantly relies on survey data and global regression models that overlook intra-urban spatial heterogeneity in shopping behaviors. This study bridges this gap by analyzing e-commerce demand’s spatial distribution from a retail service perspective, identifying key drivers, and evaluating implications for omnichannel strategies and logistics. Utilizing waybill big data, spatial analysis, and multiscale geographically weighted regression, we reveal: (1) High-density e-commerce demand areas are predominantly located in central districts, whereas peripheral regions exhibit statistically lower volumes. The spatial distribution pattern of e-commerce demand aligns with the urban development spatial structure. (2) Factors such as population density and education levels significantly influence e-commerce demand. (3) Convenience stores play a dual role as retail service providers and parcel collection points, reinforcing their importance in shaping consumer accessibility and service efficiency, particularly in underserved urban areas. (4) Supermarkets exert a substitution effect on online shopping by offering immediate product availability, highlighting their role in shaping consumer purchasing preferences and retail service strategies. These findings contribute to retail and consumer services research by demonstrating how spatial e-commerce demand patterns reflect consumer shopping preferences, the role of omnichannel retail strategies, and the competitive dynamics between e-commerce and physical retail formats. Full article