Search Results (7,778)

Frequent accidents involving heavy trucks and the inadequacy of existing dynamic monitoring technologies pose significant challenges to accurate early warning risk and safety management. To address these issues, this study proposes a multi-dimensional risk measurement and real-time early warning method for heavy truck driving behavior based on trajectory data. By extracting multi-dimensional trajectory features such as lateral position, speed, and acceleration, quantitative indicators for driving stability and car-following risk were constructed. Integrated with the CRITIC objective weighting method and the K-means++ clustering algorithm, a comprehensive risk measurement model was established to systematically characterize the dynamic evolution of driving behavior, overcoming the limitations of single-dimensional risk analysis. Experimental results based on the CQSkyEyeX trajectory dataset demonstrate that the proposed method categorizes driving behavior into six risk levels. Low-risk behavior accounted for 66.70%, while medium- to high-risk behaviors mainly included serpentine driving (26.69%) and close following (4.18%). High-risk behavior constituted only 0.03%. A multi-strategy real-time warning mechanism was further developed, achieving a warning accuracy of 98.36% with the final-value method, significantly outperforming the mode method (83.62%). The outcomes of this study demonstrate the effectiveness and practical utility of the proposed model for risk identification and early warning. On a practical level, the developed risk classification framework and management strategy establish a quantitative basis for differentiated supervision, enabling a closed-loop management process of “identification–intervention–optimization”. Future work will focus on three key directions: integrating multi-source data, extending the model to other typical operational scenarios, and incorporating advanced machine learning techniques to further enhance its generalization capability and warning accuracy. Overall, this research provides a feasible technical pathway for the precise quantification, dynamic monitoring, and tiered intervention of driving behavior in heavy-duty trucks, thereby contributing to enhanced safety in road freight transportation. Full article

Recent developments in computer vision have significantly enhanced the automation and objectivity of sports analytics. This paper proposes a novel deep learning-based framework for estimating football possession directly from broadcast video, eliminating the reliance on manual annotations or event-based data that are often labor-intensive, subjective, and temporally coarse. The framework incorporates two structurally improved object detection models: YOLOv8-P2S3A for football detection and YOLOv8-HWD3A for player detection. These models demonstrate superior accuracy compared to baseline detectors, achieving 79.4% and 71.1% validation average precision, respectively, while maintaining low computational latency. Team identification is accomplished through unsupervised DBSCAN clustering on jersey color features, enabling robust and label-free team assignment across diverse match scenarios. Object trajectories are maintained via the Norfair multi-object tracking algorithm, and a temporally aware refinement module ensures accurate estimation of ball possession durations. Extensive experiments were conducted on a dataset comprising 20 full-match Video clips. The proposed system achieved a root mean square error (RMSE) of 4.87 in possession estimation, outperforming all evaluated baselines, including YOLOv10n (RMSE: 5.12) and YOLOv11 (RMSE: 5.17), with a substantial improvement over YOLOv6n (RMSE: 12.73). These results substantiate the effectiveness of the proposed framework in enhancing the precision, efficiency, and automation of football analytics, offering practical value for coaches, analysts, and sports scientists in professional settings. Full article

The operational environment of construction machinery is predominantly unstructured, characterized by rapid changes, high complexity, and irregularly distributed objects. This poses significant challenges for 3D semantic perception, particularly due to the high cost of acquiring point cloud semantic labels. To address this, a novel 3D semantic perception scheme is proposed for such unstructured environments. This scheme integrates image semantic segmentation results with point cloud clustering via perspective projection. The projection parameters are refined using Particle Swarm Optimization (PSO), and the semantic consistency of the fused results is further enhanced by a Kd-tree-based radius nearest neighbor (RNN) matching algorithm. Consequently, a weakly supervised framework is established that achieves accurate 3D semantic understanding using only 2D image labels, eliminating the need for annotated 3D point clouds. The feasibility and effectiveness of the scheme are validated through a dedicated unstructured scene dataset and real-world testing. Results demonstrate its capability to effectively perceive 3D semantic information and reconstruct target contours, achieving a mean Pixel Accuracy (mPA) of 84.72% and a mean Intersection over Union (mIoU) of 75.85%. Full article

Traditional non-negative matrix factorization (NMF) faces challenges when dealing with complex data, primarily characterized by sensitivity to noise, neglect of data geometric structure, and inability to effectively utilize supervised information. To address these limitations, this paper proposes a class-driven robust non-negative matrix factorization with dual-hypergraph regularization (CRNMFDH) framework. The core contributions of this framework include the following: Firstly, the design of a novel dual-hypergraph regularization term that symmetrically captures and preserves the higher-order geometric structures of both the sample space and feature space, establishing a mutually reinforcing topological relationship between them. Secondly, an introduction of a class-driven mechanism to effectively integrate label information into the decomposition process, significantly enhancing the discriminative capability of the low-dimensional representations. Finally, the employment of a loss function based on correntropy to replace the traditional Euclidean distance, thereby enhancing the model’s robustness against noise and outliers. Extensive experiments across nine datasets demonstrate that CRNMFDH significantly outperforms existing state-of-the-art algorithms in multiple clustering evaluation metrics and noise robustness, providing an effective new solution for complex data clustering tasks. Full article

This study addresses the site selection problem for energy storage systems (ESSs) as a multi-criteria decision-making problem (MCDM) under conditions of uncertainty. First, potential candidate locations were identified using the K-means clustering algorithm based on the geographic coordinates of existing solar power plants (SPPs). As a result, six alternative locations representing spatial concentration were identified. These alternatives were then evaluated using the fuzzy TOPSIS method, a multi-criteria decision-making method (MCDM), taking into account the ten criteria defined for this study. Expert assessments were expressed and transformed into triangular fuzzy numbers to capture uncertainty and subjectivity in the decision-making process. The results show six alternative options, ranked from the one with the highest proximity coefficient to the one with the lowest. The findings demonstrate that the integrated use of machine learning (ML) and fuzzy TOPSIS methods provides an effective and robust decision support framework for ESS location selection problems. This approach also serves as a guide for other renewable energy planning practices. Full article

Plants defend against pathogens such as fungi by initiating coordinated structural and chemical responses. Pathogen perception triggers rapid cytosolic calcium influx and calcium oscillations that drive defense gene expression, yet the mechanisms by which these signals encode stressor intensity and propagate systematically remain unclear. Here, we present a microfluidic system to characterize intracellular calcium dynamics in protonemal colonies of the moss Physcomitrium patens (Hedw.) upon precise and reversible exposure to fungal chitin oligosaccharides. Epifluorescent imaging of cells expressing the calcium indicator GCaMP6f revealed a rapid, coordinated calcium response to chitin addition, followed by stereotyped oscillations that subsided quickly upon stimulus removal. We implemented an unbiased image segmentation algorithm using pixel-based k-means clustering to automatically locate regions with specific oscillatory signatures. Calcium dynamics were distinct across adjacent cells, distinguishable by cell type, and significantly modulated by circadian rhythm, adaptation time within the device, and stimulus timing. Cytosolic calcium oscillations, which rose and fell symmetrically within about 60 s, occurred spontaneously during the subjective night and following short adaptation periods. Chitin elicited strong oscillations with increased frequency, amplitude, and duration, and repeated pulses entrained regular, colony-wide oscillations at the stimulation interval. This study complements prior investigations of whole plant and growth tip dynamics and provides a quantitative framework to study calcium signaling in plants, including mechanisms of signal propagation and the role of oscillation frequency on gene expression. Full article

New Zealand, located along the boundary between the Pacific and Australian plates, is among the most seismically active regions in the world. In such an area, reliable short-term forecasting of strong aftershocks is essential for seismic risk mitigation. In this study, we apply NESTORE (NExt STrOng Related Earthquake), a machine learning probabilistic forecasting algorithm, to the New Zealand earthquake catalogue to evaluate the probability that a mainshock of magnitude M_m will be followed by an event of magnitude ≥ M_m − 1 within a defined space–time window. NESTORE uses nine features describing early post-mainshock seismicity and outputs the probability that a cluster is Type A (i.e., containing a strong aftershock) or not (Type B). We assess performance using two testing strategies: chronological training–testing splits and k-fold cross-validation and refine the training set using the REPENESE outlier-detection procedure. The k-fold approach proves more robust than the chronological one, despite changes in catalogue characteristics over time. Eighteen hours after the mainshock, NESTORE correctly classified 88% of clusters (75% for Type A and 92% for Type B; Precision = 0.75). Notably, the highly destructive 2010–2011 Canterbury–Christchurch sequence was correctly identified as Type A. These findings support the applicability of NESTORE for short-term aftershock forecasting in New Zealand. Full article

Background/Objectives: This study aimed to use cluster analysis of healthcare utilization patterns to identify distinct clinical phenotypes in patients with comorbid chronic obstructive pulmonary disease (COPD) and atrial fibrillation (AF) and to assess their associations with demographic characteristics and clinical outcomes. Methods: A retrospective cohort study was conducted using data from 1247 patients with COPD and AF extracted from a regional medical information system (Lipetsk Region, period 2021–2025). The k-means algorithm was used to cluster patients based on the average number of medical encounters per three-character ICD-10 categories. Groups were compared using descriptive and analytical statistical methods with correction for multiple comparisons. Results: The k-means algorithm identified three distinct clusters (phenotypes), which differed significantly in demographics, comorbidity structure, and mortality. Cluster 1 (“High-frequency utilization phenotype”, 25.3%): characterized by high utilization for acute respiratory infections, metabolic, and urological diseases; demonstrated the lowest mortality (10.1%). Cluster 2 (“Cerebrovascular Phenotype”, 32.3%): characterized by chronic cerebrovascular pathology and its sequelae (codes I67, I69); had intermediate mortality (20.8%). Cluster 3 (“Low-frequency utilization phenotype”, 42.4%): distinguished by minimal utilization for “outpatient” reasons alongside the highest mortality (31.1%) and a high proportion of deaths from respiratory failure. Analysis within the deceased patient subgroup confirmed the persistence of specific utilization patterns for each phenotype right up until the fatal outcome. Conclusions: Cluster analysis of real-world clinical practice data identified three discrete phenotypes of patients with comorbid COPD and AF, which have fundamentally different clinical–behavioral trajectories and prognoses. These findings justify the need for differentiated organizational approaches, particularly the development of proactive strategies for the active detection and engagement in follow-up care of patients with the low-frequency utilization phenotype, which is associated with the worst outcomes. Full article

Wireless Sensor Networks (WSNs) play a pivotal role in Internet of Things (IoT) applications; however, their lifetime is fundamentally constrained by the limited energy of sensor nodes. This paper introduces a Hybrid Energy-Optimized Clustering Protocol (HEOCP) that combines analytical modeling of radio energy consumption with deep learning–assisted cluster-head (CH) selection. First, an analytical framework is developed to determine the distance-constrained CH eligibility region and the optimal number of clusters, thereby minimizing redundant transmissions and balancing energy consumption. Then, a genetic algorithm (GA) is used to determine the best cluster head configuration. These configurations are then trained by a ResNet-50 deep network and averaged to reduce noise, allowing for real-time cluster head prediction without repeatedly performing expensive heuristic optimization, resulting in more steady performance. Extensive simulations under various network scales demonstrate that HEOCP extends network lifetime by up to 60% compared with conventional LEACH and GA-based approaches, effectively delaying the first-node death and improving overall energy efficiency. Furthermore, the hybrid GA–ResNet framework exhibits high scalability and computational efficiency, making it suitable for large-scale IoT deployments. The results confirm that integrating analytical energy modeling with deep learning provides a powerful and sustainable paradigm for intelligent energy management in future IoT-enabled WSNs. Full article

Edge computing has garnered significant attention in recent years due to its potential in distributed systems. However, the dynamic and heterogeneous nature of edge environments introduces substantial challenges for task scheduling. Conventional rule-based scheduling algorithms often fail to adapt to rapid load fluctuations, resulting in cluster load imbalance and suboptimal resource utilization. To address this issue, we propose a container-based edge cluster scheduling framework designed to enhance load balancing. Within this framework, we introduce an Experience-Prioritized Reinforcement Scheduler (EPRS), which leverages a priority-driven sample selection mechanism to facilitate focused learning of high-value samples. The EPRS dynamically monitors node resource states via a real-time resource monitor and optimizes multi-dimensional resource allocation by jointly considering node-level metrics (e.g., computational resources, memory pressure, storage performance, and container density) and task-specific resource requirements. To validate our approach, we implemented a system prototype integrated with the proposed framework and EPRS in a Kubernetes-based edge cluster. Experimental results demonstrate that the proposed method significantly improves multi-dimensional load balancing performance, achieving an average gain of 28.25% over existing reinforcement learning-based scheduling approaches and a 29.78% improvement compared with the traditional scheduling algorithm. Full article

This research explores the development of a decoder-only speech language model (SLM) for Kazakh, a language currently characterized by limited computational resources. Our approach leverages discrete acoustic units synthesized from self-supervised speech representations. Specifically, we utilize a pretrained Wav2Vec 2.0 model to extract continuous latent features, which are then transformed into discrete semantic tokens via the k-means clustering algorithm. These tokens serve as the foundation for training a generative model designed to predict and maximize the likelihood of speech-unit sequences. To facilitate this study, we curated a specialized Kazakh speech corpus by synthesizing and refining multiple publicly available audio datasets. Given the constrained hardware resources available, we conducted large-scale feature extraction and tokenization to train the unit-based model. We evaluated the system’s efficacy using negative log-likelihood and perplexity metrics on independent test sets. The model captures Kazakh vowel harmony but struggles with long-range agglutinative chains. Key observations include the model’s high sensitivity to data quality, tokenization techniques, and specific training hyperparameters. Although constrained by data volume and training time relative to global benchmarks, the model successfully captures the underlying structural patterns in Kazakh speech. This work establishes a vital empirical baseline and suggests future improvements through refined unit discovery and integrated speech-text modeling. Full article

Along with the swift evolution of autonomous driving and internet technologies, In-Vehicle Ethernet has evolved into the core backbone network underpinning the new generation of in-vehicle networks (IVNs). Since In-Vehicle Ethernet is susceptible to a host of cybersecurity threats—such as data pilferage, data falsification, and malicious unauthorized access—it is imperative to enhance its defense capabilities. This research focuses on anomaly identification for In-Vehicle Ethernet communication networks, with a specific focus on the intrinsic data features of the AVTP protocol and potential cyber-attack vectors targeting the network. This work develops a novel network anomaly detection approach rooted in the Fuzzy clustering algorithm. This effectively enhances the cybersecurity performance of In-Vehicle Ethernet. Experimental results demonstrate that the Fuzzy clustering algorithm proposed in this study achieves 97.4% accuracy in detecting anomalous data, outperforming the traditional K-Means and OPTICS clustering algorithms by 6.4% and 14.5% respectively in anomaly detection rate. This further elevates the cybersecurity performance of In-Vehicle Ethernet and forges a robust foundation for the stable operation and iterative advancement of intelligent connected vehicles (ICVs). Full article

Wind-solar power generation is inherently uncertain. These uncertainties bring considerable difficulties to the assessment of hosting capacity. To tackle these difficulties, it is essential to create typical scenarios that can precisely capture the statistical traits and interrelationships of wind-solar power. In this research, we systematically integrate various scenario generation techniques, resulting in the creation of a holistic framework grounded in kernel density estimation (KDE) and Copula functions. Our proposed approach represents the stochastic nature of wind-solar power output by constructing their respective probability density functions (PDFs). It comprehensively depicts the potential spatiotemporal complementarity between wind-solar power by utilizing Copula functions and establishing a joint probability distribution model. Through Monte Carlo simulation, we generated a large number of wind-solar output scenarios. Subsequently, we employed the K-means clustering algorithm to reduce the number of scenarios. The findings reveal that the integrated framework, which combines KDE and Copula theory, achieves higher fitting accuracy for the marginal distributions and correlation structures of wind-solar power generation. As a result, the generated scenarios are more representative and reliable, offering strong support for photovoltaic (PV) hosting capacity analysis (HCA) and the formulation of typical plans. We validate the proposed method using historical wind-solar data from several representative regions in China, such as Inner Mongolia, northern Hebei, the Beijing–Tianjin–Hebei region, and Hubei Province. This validation demonstrates the method’s applicability under various geographical and climatic conditions. Full article

This article examines the revolutionary impact of Artificial Intelligence (AI) on transforming cryptocurrency trading, a sector characterised by extreme volatility, dynamism, and nonlinear data. Through a rigorous bibliometric analysis based on the Web of Science database, this study examines a sample of 555 scientific papers published between 2016 and 2025, utilising the PRISMA protocol for systematic selection, and tools such as VOSviewer and MS Excel. The analysis identifies five major thematic clusters: (1) blockchain infrastructure and AI integration in decentralised ecosystems, (2) data analysis and practical applicability in crypto markets, (3) financial and social data analysis—machine learning algorithms, (4) algorithmic trading and automation, and (5) prediction and modelling of crypto market developments. The originality of this study lies in providing an overview of the implementation stage of these technologies by integrating the results into a map of Technology Readiness Levels (TRLs). The findings highlight a clear transition from traditional statistical methods to autonomous decision-making systems capable of processing massive volumes of data for portfolio optimisation. This study’s limitation is that it may require periodic updates, as the AI and cryptocurrency landscape are constantly evolving. Full article

To enhance the resilience and sustainability of urban metro systems under operational uncertainties and external disturbances, critical station identification and vulnerability assessment should be further investigated from the perspective of network science. In this paper, the presented comprehensive clustering algorithm and the Pearson correlation coefficient are adopted to explore the origin-destination (OD) passenger flow characteristics on different date classifications, and the different dates should be reasonably classified into three categories, including working day, weekends, and holiday. Meanwhile, this paper proposes the dynamic DomiRank algorithm and flow DomiGCN model to identify critical stations from network structure and function on different data classifications respectively, and further studies the vulnerability property of metro networks under simulated attacks. The Shanghai metro network is selected as case to prove the feasibility and correctness of the model. The results show that the dynamic DomiRank algorithm is relatively effective to identify critical stations from network structure, and the flow DomiGCN model is also relatively effective to identify critical stations from network function. Moreover, simulated attacks to these critical stations detected by the proposed methods can cause more damages than the other methods. These findings provide some supports for protection of metro infrastructure and contribute to the sustainable operation and development of urban rail transit systems. Full article