Search Results (5,024)

Existing unsupervised anomaly detection methods suffer from insufficient parameter precision, poor robustness to noise, and limited generalization capability. To address these issues, this paper proposes an Adaptive Diffusion Adversarial Evolutionary Network (ADAEN) for unsupervised anomaly detection in tabular data. The proposed network employs an adaptive hierarchical feature evolution generator that captures multi-scale feature representations at different abstraction levels through learnable attribute encoding and a three-layer Transformer encoder, effectively mitigating the gradient vanishing problem and the difficulty of modeling complex feature relationships that are commonly observed in conventional generators. ADAEN incorporates a multi-scale adaptive diffusion-augmented discriminator, which preserves scale-specific features across different diffusion stages via cosine-scheduled adaptive noise injection, thereby endowing the discriminator with diffusion-stage awareness. Furthermore, ADAEN introduces a multi-scale robust adversarial gradient loss function that ensures training stability through a diffusion-step-conditional Wasserstein loss combined with gradient penalty. The method has been evaluated on 14 UCI benchmark datasets and achieves state-of-the-art performance in anomaly detection compared to existing advanced algorithms, with an average improvement of 8.3% in AUC, an 11.2% increase in F1-Score, and a 15.7% reduction in false positive rate. Full article

As the “brain” of the information industry and modern manufacturing, chips have emerged as a focal point in global competition over critical technologies. Based on global chip trade data from 2010 to 2023, this study employs social network analysis to investigate the structural evolution of the chip trade network and applies the quadratic assignment procedure (QAP) to examine the driving mechanisms of network reconstruction. The findings are as follows: First, the global chip trade network exhibits a loosely connected core-periphery structure, characterized by clustering and polarization, with a pronounced short-term deglobalization trend. Second, China, the United States, Germany, France, South Korea, and Singapore have long dominated central positions in competitive dynamics, while developing economies such as Mexico, Malaysia, and the Philippines have significantly risen in prominence in recent years. Third, the network takes on a core–subcore–periphery configuration with clearly delineated trade communities, reflecting a community-based, multi-centric, and hierarchical pattern. Fourth, political relations serve as a key driver of network restructuring, with their promotional effect on chip trade being negatively moderated by technological distance yet positively moderated by economic-complexity distance. Full article

Lumbar shear forces are increasingly recognized as critical contributors to lower-back injury risk, yet most ergonomic assessment tools—most notably the Revised NIOSH Lifting Equation (RNLE)—do not directly estimate shear loading. This study develops and evaluates a family of linear mixed-effects regression models that statistically predict L4/L5 lumbar shear force exposure using traditional NIOSH lifting parameters combined with posture descriptors extracted from digital human models. A harmonized dataset of 106 peak-shear lifting postures was compiled from five controlled laboratory studies, with lumbar shear forces obtained from validated biomechanical simulations implemented in the Siemens JACK (Siemens software, Plano, TX, USA) platform. Twelve model formulations were examined, varying in fixed-effect structure and hierarchical random effects, to quantify how load magnitude, hand location, sex, and joint posture relate to simulated task-level anterior–posterior shear exposure at the lumbar spine. Across all models, load magnitude and horizontal reach emerged as the strongest and most stable predictors of shear exposure, reflecting their direct mechanical influence on anterior spinal loading. Hip and knee flexion provided substantial additional explanatory power, highlighting the role of whole-body posture strategy in modulating shear demand. Upper-limb posture and coupling quality exhibited minimal or inconsistent effects once load geometry and lower-body posture were accounted for. Random-effects analyses demonstrated that meaningful variability arises from individual movement strategies and task conditions, underscoring the necessity of mixed-effects modeling for representing hierarchical structure in lifting data. Parsimonious models incorporating subject-level random intercepts produced the most stable and interpretable coefficients while maintaining strong goodness-of-fit. Overall, the findings extend the NIOSH framework by identifying posture-dependent determinants of lumbar shear exposure and by demonstrating that simulated shear loading can be reliably predicted using ergonomically accessible task descriptors. The proposed models are intended as statistical predictors of task-level shear exposure that complement—rather than replace—comprehensive biomechanical simulations. This work provides a quantitative foundation for integrating shear-aware metrics into ergonomic risk assessment practices, supporting posture-informed screening of manual material-handling tasks in field and sensor-based applications. Full article

With the rapid development and widespread application of multibeam echo-sounding systems, large-scale and high-resolution seafloor topography can be efficiently acquired, enabling precise mapping of seabed terrain. However, due to complex oceanographic conditions, instrumental noise, and acoustic interferences, the acquired multibeam data often contain outliers that deviate from the true seafloor surface. These outliers can distort the representation of seafloor topography, adversely affecting subsequent geological analysis and engineering applications. To address this issue, a hybrid outlier detection method combining CUBE filtering with the Isolation Forest (IForest) algorithm, termed CUBE-IForest, is proposed. The method first employs CUBE filtering to remove gross outliers based on local uncertainty estimation, followed by the application of IForest to identify subtle anomalies in the refined data, achieving hierarchical detection of outliers. Experimental results based on in situ multibeam bathymetric data from the northeastern Pacific demonstrate that compared with traditional filtering methods the CUBE-IForest approach significantly improves detection accuracy and reduces both false positive and false negative rates by approximately 30%, confirming its efficiency and reliability in seafloor mapping and analysis. Full article

In today’s corporate environment, employees’ proactive engagement in environmental behaviors is crucial for the effective implementation of corporate environmental regulations. Leadership is crucial in motivating such behaviors. This study, grounded in self-determination theory, explores how Environmental Responsible Leadership enhances employees’ Organizational Citizenship Behavior for the Environment through Psychological Empowerment. Additionally, we analyze the moderating role of Green Culture. Using a multi-wave survey design, data were collected from 262 corporate employees in China via the Credamo platform and analyzed through structural equation modeling (SEM) with AMOS 24.0 and hierarchical regression analysis with SPSS 26.0. The results reveal that Environmental Responsible Leadership significantly promotes the enhancement in Organizational Citizenship Behavior for the Environment. Psychological Empowerment serves as a significant mediator in this relationship, while Green Culture, as a supportive organizational context, amplifies the positive effects of Environmental Responsible Leadership on employees’ environmental behaviors. By elucidating the mechanisms and boundary conditions of Environmental Responsible Leadership, this study provides practical insights for organizations seeking to advance ecological conservation through leadership development and cultural nurturing. Full article

Large Language Models (LLMs) have demonstrated exceptional performance in knowledge-based dialogue generation and text evaluation. Synthetic data serves as a cost-effective alternative for generating high-quality datasets. However, it often plagued by hallucinations, inconsistencies, and self-anthropomorphized responses. Concurrently, manual construction of knowledge-based dialogue datasets remains bottlenecked by prohibitive costs and inherent human subjectivity. To address these multifaceted challenges, we propose ACE (Automatic Construction of Knowledge-Grounded and Engaging Human–AI Conversation Dataset), a hybrid method using hierarchical prompt engineering. This approach mitigates hallucinations and self-personalization while maintaining response consistency. Furthermore, existing human and automated evaluation methods struggle to assess critical factors like factual accuracy and coherence. To overcome this, we introduce the Truthful Answer Score (TAS), a novel metric specifically designed for knowledge-based dialogue evaluation. Our experimental results demonstrate that the ACE dataset achieves higher quality than existing benchmarks, such as Wizard of Wikipedia (WoW) and FaithDial. Additionally, TAS aligns more closely with human judgment, offering a more reliable and scalable evaluation framework. Our findings demonstrate that leveraging LLMs through systematic prompting can substantially reduce reliance on human annotation while simultaneously elevating the quality and reliability of synthetic datasets. Full article

Background: Sickle cell disease (SCD) is a prevalent hereditary hemoglobinopathy associated with substantial morbidity, particularly in regions with limited access to advanced laboratory diagnostics. Conventional diagnostic workflows, including manual peripheral blood smear examination and biochemical or molecular assays, are resource-intensive, time-consuming, and subject to observer variability. Recent advances in artificial intelligence (AI) enable automated analysis of blood smear images and offer a scalable alternative for SCD screening. Methods: This study presents a controlled benchmark of CNNs, Vision Transformers, hierarchical Transformers, and hybrid CNN–Transformer architectures for image-level SCD classification using a publicly available peripheral blood smear dataset. Eleven ImageNet-pretrained models were fine-tuned under identical conditions using an explicit leakage-safe evaluation protocol, incorporating duplicate-aware, group-based data splitting and repeated splits to assess robustness. Performance was evaluated using accuracy and macro-averaged precision, recall, and F1-score, complemented by bootstrap confidence intervals, paired statistical testing, error-type analysis, and explainable AI (XAI). Results: Across repeated group-aware splits, CNN-based and hybrid architectures demonstrated more stable and consistently higher performance than transformer-only models. MaxViT-Tiny and DenseNet121 ranked highest overall, while pure ViTs showed reduced effectiveness under data-constrained conditions. Error analysis revealed a dominance of false-positive predictions, reflecting intrinsic morphological ambiguity in challenging samples. XAI visualizations suggest that CNNs focus on localized red blood cell morphology, whereas hybrid models integrate both local and contextual cues. Conclusions: Under limited-data conditions, convolutional inductive bias remains critical for robust blood-smear-based SCD classification. CNN and hybrid CNN–Transformer models offer interpretable and reliable performance, supporting their potential role as decision-support tools in screening-oriented research settings. Full article

With the wide application of cloud computing in multi-tenant, heterogeneous nodes and high-concurrency environments, model parameters frequently interact during distributed training, which easily leads to privacy leakage, communication redundancy, and decreased aggregation efficiency. To realize the collaborative optimization of privacy protection and computing performance, this study proposes the Heterogeneous Federated Homomorphic Encryption Cloud (HFHE-Cloud) model, which integrates federated learning (FL) and homomorphic encryption and constructs a secure and efficient collaborative learning framework for cloud platforms. Under the condition of not exposing the original data, the model effectively reduces the performance bottleneck caused by encryption calculation and communication delay through hierarchical key mapping and dynamic scheduling mechanism of heterogeneous nodes. The experimental results show that HFHE-Cloud is significantly superior to Federated Averaging (FedAvg), Federated Proximal (FedProx), Federated Personalization (FedPer) and Federated Normalized Averaging (FedNova) in comprehensive performance, Homomorphically Encrypted Federated Averaging (HE-FedAvg) and other five baseline models. In the dimension of privacy protection, the global accuracy is up to 94.25%, and the Loss is stable within 0.09. In terms of computing performance, the encryption and decryption time is shortened by about one third, and the encryption overhead is controlled at 13%. In terms of distributed training efficiency, the number of communication rounds is reduced by about one fifth, and the node participation rate is stable at over 90%. The results verify the model’s ability to achieve high security and high scalability in multi-tenant environment. This study aims to provide cloud service providers and enterprise data holders with a technical solution of high-intensity privacy protection and efficient collaborative training that can be deployed in real cloud platforms. Full article

This study addresses the critical problem of barriers hindering Green Supply Chain Management (GSCM) adoption in Egypt’s petrochemical sector, a major economic driver that produces approximately 4.5 million tons annually but generates significant GHG emissions and hazardous waste. The objective is to identify, rank, and analyze the hierarchical relationships among internal and external barriers using a mixed-methods approach. This study focuses on the full petrochemical supply chain in Egypt, encompassing upstream (raw material sourcing), midstream (manufacturing/refining processes), and downstream (distribution, waste management, reverse logistics), with an emphasis on emission/waste reduction practices. Data were collected via a structured questionnaire from 400 employees in Egyptian petrochemical firms and analyzed using Interpretive Structural Modeling (ISM). The findings showed that internal impediments, such as a lack of corporate leadership and support (IB1), a critical shortage of resources (IB6), and the absence of green initiatives (IB5), serve as driving forces that exert a cascading influence over the external barriers, which include insufficient government support (EB1), a lack of markets for recycled materials (EB5), and human resources or expertise shortages (EB7). The study contributes to the existing literature on GSCM by incorporating international trends and specifically addressing Egyptian issues, including weak policies, difficult supply chains, high energy-intensive operations, and costly operations. The study suggests that sending clear messages from the top and providing financial incentives can help push the obstacles aside and guide the industry down the path of environmentally responsible operations. Full article

Crowdfunding has emerged as an alternative funding source among entrepreneurs, businesses, and industries. In recent years, research on machine learning-based project classification models has been conducted with the aim of predicting the success of crowdfunding campaigns, both for entrepreneurs and investors. However, most of the research has focused on classification approaches using non-content information such as project metadata, creators’ behavior, and social history, but there have been few attempts to use text content data per se, particularly in order to provide explanations and evidence for how the prediction decisions were made. To address this point, we propose to use a deep learning-based approach called Hierarchical Attention Network (HAN) to predict the success of crowdfunding campaigns and provide explanation and justification of the prediction decisions using attention weights. We collect publicly available data of crowdfunding campaigns and build our success prediction model with an accuracy of 86.38% and 87.29%, using an Updates section and backers’ comments in a Comments section, respectively. We also explore the feasibility of early success prediction during the funding period (up to 2 months), with as much as 80.99% accuracy in 1 to 2 months. Finally, we examine word and sentence attention weight scores to clarify key factors in predicting crowdfunding success. Full article

Urbanization in mountainous regions alters hydrologic systems, yet the spatial patterning of residential (RA) and non-residential (NRA) areas in response to hydrologic constraints remains poorly quantified. In this study, we analyzed how such constraints shaped the distinct locational logic of RA and NRA expansion in the mountainous Kunming Core Region (KCR), Southwest China, from 1975 to 2020. Using the Global Human Settlement Layer (GHS-BUILT-S) built-up fraction data and its functionally classified RA and NRA layers at 100 m resolution, we quantified multi-decadal urban land changes via regression and centroid migration analyses. Six hydrologic factors, namely altitude, slope, surface roughness, distance to river (DTR), depth to water table (DTWT) and vegetation root depth (VRD), were derived from global terrain, groundwater, and rooting depth datasets, and harmonized to a common grid. Results show a two-phase urbanization pattern: moderate, compact growth before 1995 followed by rapid, near-exponential expansion, dominated by RA. RA consistently clustered in hydrologically favorable zones (low–moderate roughness, mid-altitudes, lower slopes, proximal rivers, shallow–moderate DTWT, moderate VRD), whereas NRA expanded into more hydrologically variable terrain (higher roughness, intermediate DTR, deeper DTWT, higher altitudes, deeper VRD). Contribution-weighting analysis revealed a temporal shift in dominant drivers: for RA, from river proximity and slope in 1975 to terrain roughness in 2020; for NRA, from vegetation root depth and moderate topography to root depth plus altitude. Geographic centroids of both RA and NRA migrated northeastward, indicating coordinated yet functionally distinct peri-urban and corridor-oriented growth. These findings provide a hierarchical, factor-based framework for integrating hydrologic constraints into risk-informed land-use planning in topographically complex basins. Full article

The continuous handling of the large amount of raw data generated by implantable brain–computer interface (BCI) devices requires a large amount of hardware resources and is becoming a bottleneck for implantable BCI systems, particularly for power-constrained wireless systems. To overcome this bottleneck, we present NeuroGator, an asynchronous gating system using Local Field Potential (LFP) for the implantable BCI system. Unlike a conventional continuous data decoding approach, NeuroGator uses hierarchical state classification to efficiently allocate hardware resources to reduce the data size before handling or transmission. The proposed NeuroGator operates in two stages: Firstly, a low-power hardware silence detector filters out background noise and non-active signals, effectively reducing the data size by approximately $69.4\%$. Secondly, a Dual-Resolution Gate Recurrent Unit (GRU) model controls the main data processing procedure on the edge side, using a first-level model to scan low-precision LFP data for potential activity and a second-level model to analyze high-precision LFP data for confirmation of an active state. The experiment shows that NeuroGator reduces overall data throughput by 82% while maintaining an F1-Score of 0.95. This architecture allows the Implantable BCI system to stay in an ultra-low-power state for over 85% of its entire operation period. The proposed NeuroGator has been implemented in an Application-Specific Integrated Circuit (ASIC) with a standard 180 nm Complementary Metal Oxide Semiconductor (CMOS) process, occupying a silicon area of $0.006{\mathrm{mm}}^2$ and consuming 51 nW power. NeuroGator effectively resolves the resource efficiency dilemma for implantable BCI devices, offering a robust paradigm for next-generation asynchronous implantable BCI systems. Full article

LNG serves as a pivotal element within integrated energy systems, especially in coastal regions where the implementation of a stable and reliable LNG cold energy power generation system significantly elevates energy efficiency. This system can effectively meet concurrent demands for cold energy utilization and electricity supply while contributing to the mitigation of carbon emissions. However, the inherent complexity of the system coupled with the scarcity of historical operational data for the novel dual-Rankine cycle process renders conventional reliability assessment methodologies inadequate. This study proposes an integrated framework utilizing fuzzy Bayesian methods to address data scarcity during the early stages of equipment deployment. A hierarchical risk factor model, incorporating process decomposition, expert evaluations, and triangular fuzzy numbers, is developed to quantify uncertainties in failure probabilities. The Bayesian network models the causal relationships among equipment failure factors, allowing for the inference of overall system reliability from individual equipment performance. Through a case study of a LNG terminal in Zhoushan, this approach integrates sensitivity analysis with forward-backward reasoning methodologies to rigorously evaluate and quantify system reliability under operational conditions. The results show that under high load conditions within the 1000 h prior to overhaul, following long-term accumulated operation, the probability of complete system shutdown in the power generation system is 3.30%, while the probability of the LNG cold energy power generation system failing to operate fully due to aging-related faults is 8.24%, demonstrating the system’s strong reliability under extreme conditions. Critical risks identified through backward inference include the seawater pump SWP1, with a posterior failure probability of 59.92% during complete shutdown, and the propane-side pump SWP3, with a posterior failure probability of 32.29% when the cold energy power generation system can only operate in a single-cycle mode. This study provides an advanced methodological framework for risk management in newly constructed LNG cold energy power generation systems, playing a crucial role in promoting sustainable, low-carbon technologies in the energy sector. Full article

Distributed photovoltaic (DPV) systems present a cost-effective and sustainable industrial energy solution, yet their reliable monitoring faces significant technological constraints. This paper proposes a hierarchical optimization framework that integrates hysteresis-based traffic shaping at the network layer with Large Language Model (LLM)-driven diagnostics at the application layer. The proposed dynamic algorithm minimizes latency and downtime by prioritizing critical fault data. Priority-based scheduling ensures this critical data is transmitted preferentially over routine sensor readings. At the application layer, the system utilizes physics-informed prompt engineering to perform zero-shot root cause analysis, circumventing the training data requirements of traditional classifiers. Under a 10 Mbps gateway bandwidth, our method achieves a 46.08% to 49.87% reduction in P50 latency compared to traditional approaches. Moreover, the LLM-powered diagnostic system provides detailed assessments, enabling precise fault diagnosis for DPV systems. Full article

Background/Objectives: Lesbian, gay, bisexual, and transgender (LGBT) cancer survivors face disproportionately high structural and psychosocial barriers to post-diagnosis care. However, heterogeneity within this population remains understudied. This study aimed to characterize healthcare utilization (HCU) barriers among LGBT cancer survivors, assess psychosocial vulnerabilities (discrimination, stress, and social support), and identify survivor subgroups at greatest risk for care disengagement. Methods: Data were drawn from the All of Us Research Program. A sample of 3502 LGBT cancer survivors was analyzed, including lesbian (n = 730), gay (n = 1285), bisexual (n = 1296), and transgender/gender expansive (TGE) (n = 209) individuals. HCU barriers were assessed using 21 binary indicators. Psychosocial measures included the Everyday Discrimination Scale, Perceived Stress Scale, and MOS Social Support Survey. Agglomerative hierarchical cluster analysis identified latent HCU barrier profiles. Differences across clusters and identity groups were assessed using ANOVA and chi-square tests, and multinomial logistic regression examined demographics, socioeconomic, and psychosocial predictors of cluster membership. Results: Three distinct HCU barrier clusters were identified: low (59.7%), moderate (27.8%), and high (12.5%). Bisexual and TGE survivors were disproportionately represented in the high-barrier cluster, which was characterized by widespread cost-related nonadherence, structural delays in care, and higher levels of perceived discrimination and stress. In adjusted models, bisexual identity, lower income, female sex assigned at birth, and higher discrimination and perceived stress were independently associated with increased odds of high-barrier cluster membership. Conclusions: Substantial heterogeneity exists in HCU barriers among LGBT cancer survivors. Bisexual and TGE survivors experience a concentrated burden of structural and psychosocial barriers to survivorship care, highlighting the relevance of targeted, data-driven approaches to reduce access inequities within this population. Full article