Search Results (357)

Amid the increasing frequency and severity of major disasters, the rapid spread of online misinformation poses substantial risks to public safety, effective crisis management, and long-term societal sustainability. Current methods for managing disaster-related rumors rely on static, rule-based approaches that lack scalability, fail to capture nuanced misinformation, and are limited to reactive responses, hindering effective disaster management. To address this gap, this study proposes a novel framework that leverages large language models (LLMs) and event knowledge graphs (EKGs) to facilitate the sustainable agile identification and adaptive control of disaster-related online rumors. The framework follows a multi-stage process, which includes the collection and preprocessing of disaster-related online data, the application of Gaussian Mixture Wasserstein Autoencoders (GMWAEs) for sentiment and rumor analysis, and the development of EKGs to enrich the understanding and reasoning of disaster events. Additionally, an enhanced model for rumor identification and risk control is introduced, utilizing Graph Attention Networks (GATs) to extract node features for accurate rumor detection and prediction of rumor propagation paths. Extensive experimental validation confirms the efficacy of the proposed methodology in improving disaster response. This study contributes novel theoretical insights and presents practical, scalable solutions for rumor control and risk management during crises. Full article

Accurate prediction of the RUL of electric drive bearings over the entire service life cycle for new energy vehicles optimizes maintenance strategies and reduces costs, addressing clear application needs. Full life data of electric drive bearings exhibit long time spans and abrupt degradation, complicating the modeling of time dependent relationships and degradation states; therefore, a piecewise linear degradation model is appropriate. An RUL prediction method is proposed based on degradation assessment and spatiotemporal feature fusion, which extracts strongly time correlated features from bearing vibration data, evaluates sensitive indicators, constructs weighted fused degradation features, and identifies abrupt degradation points. On this basis, a piecewise linear degradation model is constructed that uses a path graph structure to represent temporal dependencies and a temporal observation window to embed temporal features. By incorporating GAT-LSTM, RUL prediction for bearings is performed. The method is validated on the XJTU-SY dataset and on a loaded ball bearing test rig for electric vehicle drive motors, yielding comprehensive vibration measurements for life prediction. The results show that the method captures deep degradation information across the full bearing life cycle and delivers accurate, robust predictions, providing guidance for the health assessment of electric drive bearings in new energy vehicles. Full article

Credit card fraud detection remains a major challenge due to severe class imbalance and the constantly evolving nature of fraudulent behaviors. To address these challenges, this paper proposes a hybrid framework that integrates a Variational Autoencoder (VAE) for probabilistic anomaly detection, a Graph Attention Network (GAT) for capturing inter-transaction relationships, and a stacking ensemble with XGBoost for robust prediction. The joint use of VAE anomaly scores and GAT-derived node embeddings enables the model to capture both feature-level irregularities and relational fraud patterns. Experiments on the European Credit Card and IEEE-CIS Fraud Detection datasets show that the proposed approach outperforms baseline models by up to 15% in F1-score, achieving values above 0.980 with AUCs reaching 0.995. These results demonstrate the effectiveness of combining unsupervised anomaly detection with graph-based learning within an ensemble framework for highly imbalanced fraud detection problems. Full article

As an emerging distributed energy system, microgrid power flow prediction plays a crucial role in optimizing energy dispatch and power grid operation. Traditional methods of power flow prediction mainly rely on statistics and time series models, neglecting the spatial relationships among different nodes within the microgrid. To overcome this limitation, a Physical-Informed Graph Attention Network (PI-GAT) is proposed to capture the spatial structure of the microgrid, while an attention mechanism is introduced to measure the importance weights between nodes. In this study, we constructed a representative 14-node microgrid power flow dataset. After collecting the data, we preprocessed and transformed it into a suitable format for graph neural networks. Next, an autoencoder was employed for pre-training, enabling unsupervised learning-based dimensionality reduction to enhance the expressive power of the data. Subsequently, the extended data is fed into a graph convolution module with attention mechanism, allowing adaptive weight learning and capturing relationships between nodes. And integrate the physical state equation into the loss function to achieve high-precision power flow prediction. Finally, simulation verification was conducted, comparing the PI-GAT method with traditional approaches. The results indicate that the proposed model outperforms the other latest model across various evaluation indicators. Specifically, it has 46.9% improvement in MSE and 14.08% improvement in MAE. Full article

Modern fire alarm systems are essential for public safety, yet they often fail to exploit the wealth of historical alarm data and the complex spatiotemporal dependencies inherent in urban environments. Graph Neural Networks (GNNs) are currently among the most popular methods for handling complex spatiotemporal dependencies. While a range of dynamic GNN approaches have been proposed, many existing GNN-based predictors still rely on a static topology, which limits their ability to fully capture the evolving nature of risk propagation. Furthermore, even among dynamic graph methods, most focus on temporal link prediction or social interaction modeling, with limited exploration in safety-critical applications such as fire alarm prediction. DeaGAT dynamically updates inter-building edge weights through an attention mechanism, enabling the graph structure to evolve in response to shifting risk patterns. A margin-based contrastive learning objective further enhances the quality of node embeddings by distinguishing subtle differences in risk states. In addition, DeaGAT jointly models static building attributes and dynamic alarm sequences, effectively integrating long-term semantic context with short-term temporal dynamics. Extensive experiments on real-world datasets, including comparisons with state-of-the-art baselines and comprehensive ablation studies, demonstrate that DeaGAT achieves superior accuracy and F1-score, validating the effectiveness of dynamic graph updating and contrastive learning in enhancing proactive fire early-warning capabilities. Full article

Trajectory planning for logistics UAVs in complex environments faces a key challenge: balancing global search breadth with fine constraint accuracy. Traditional algorithms struggle to simultaneously manage large-scale exploration and complex constraints, and lack sufficient modeling capabilities for multi-UAV systems, limiting cluster logistics efficiency. To address these issues, we propose a GWOP algorithm based on dual-layer fusion of GWO and GRPO and incorporate a graph attention network (GAT). First, CEC2017 benchmark functions evaluate GWOP convergence accuracy and balanced exploration in multi-peak, high-dimensional environments. A hierarchical collaborative architecture, “GWO global coarse-grained search + GRPO local fine-tuning”, is used to overcome the limitations of single-algorithm frameworks. The GAT model constructs a dynamic “environment–UAV–task” association network, enabling environmental feature quantification and multi-constraint adaptation. A multi-factor objective function and constraints are integrated with multi-task cascading decoupling optimization to form a closed-loop collaborative optimization framework. Experimental results show that in single UAV scenarios, GWOP reduces flight cost (FV) by over 15.85% on average. In multi-UAV collaborative scenarios, average path length (APL), optimal path length (OPL), and FV are reduced by 4.08%, 14.08%, and 24.73%, respectively. In conclusion, the proposed method outperforms traditional approaches in path length, obstacle avoidance, and trajectory smoothness, offering a more efficient planning solution for smart logistics. Full article

Cooperative Multi-Agent Reinforcement Learning (MARL) relies on effective coordination among agents to maximize team performance in complex environments. However, existing coordination graph-based approaches often overlook dynamic group structures and struggle to accurately capture fine-grained inter-agent dependencies. In this paper, we introduce a novel method called the Group Attention Aware Coordination Graph (G2ACG), which builds upon the group modeling capabilities of the Group-Aware Coordination Graph (GACG). G2ACG incorporates a dynamic attention mechanism to dynamically compute edge weights in the coordination graph, enabling a more flexible and fine-grained representation of agent interactions. These learned edge weights guide a Graph Attention Network (GAT) to perform message passing and representation learning, and the resulting features are integrated into a global policy via QMIX for cooperative decision-making. Experimental results on the StarCraft II Multi-Agent Challenge (SMAC) benchmark show that G2ACG consistently outperforms strong baselines, including QMIX, DICG, and GACG, across various scenarios with diverse agent types and population sizes. Ablation studies further confirm the effectiveness of the proposed attention mechanism, demonstrating that both the number of attention heads and the number of GAT layers significantly affect performance, with a two-layer GAT and multi-head attention configuration yielding the best results. Full article

To enhance the efficiency and reduce the operational costs of large-scale Unmanned Aerial Vehicle (UAV) inspection missions limited by endurance, this paper addresses the coupled problem of dynamically positioning landing/takeoff sites and routing the UAVs. A novel Hierarchical Reinforcement Learning (H-DRL) framework is proposed, which decouples the problem into a high-level strategic deployment policy and a low-level tactical routing policy. The primary contribution of this work lies in two architectural innovations that enable globally coordinated, end-to-end optimization. First, a coordinated credit assignment mechanism is introduced, where the high-level policy communicates its strategic guidance to the low-level policy via a learned “intent vector,” facilitating intelligent collaboration. Second, an Energy-Aware Graph Attention Network (Ea-GAT) is designed for the low-level policy. By endogenously embedding an energy feasibility model into its attention mechanism, the Ea-GAT guarantees the generation of dynamically feasible flight paths. Comprehensive simulations and a physical experiment validate the proposed framework. The results demonstrate a significant improvement in mission efficiency, with the makespan reduced by up to 16.3%. This work highlights the substantial benefits of joint optimization for dynamic robotic applications. Full article

This work addresses the broken symmetry in syntactic–semantic representations for Aspect-Based Sentiment Analysis, where advancements have been driven by the use of pre-trained language models to achieve contextual understanding and graph neural networks capturing aspect–opinion dependencies using syntactic trees. However, long-distance aspect–opinion pairs pose challenges: the structural noise in dependency trees often causes erroneous associations, while the discrete structure of the constituent trees leads to constituent fragmentation. In this paper, we propose DySynGAT and introduce a Localized Graph Attention Network (LGAT) to fuse bi-gram syntactic and semantic information from both dependency and constituent trees, effectively mitigating interference from dependency tree noise. A dynamic semantic enhancement module efficiently integrates local and global semantics, alleviating constituent fragmentation caused by constituent trees. An aspect–context interaction graph (ACIG), built upon minimal semantic segmentation and jointly enhanced features, filters out noisy cross-clause edges. Spatial reduction attention (SRA) with mean pooling compresses the redundant sequential features, reducing the noise under long-range dependencies. Experiments on foods and beverages, electronics, and user review datasets demonstrate F1 score improvements of 0.55%, 3.55%, and 1.75% over SAGAT-BERT, demonstrating strong cross-domain robustness. Full article

Arabinogalactan proteins (AGPs) and extensins influence cell wall assembly and regulate plant cell mechanical properties through interactions with extracellular matrix polymers. These proteins may play a key role in the biochemical events underlying postharvest treatments aimed at controlling fruit texture and turgor loss associated with senescence-related disorders. We studied the temporal and spatial accumulation patterns of extensin and AGP isoforms constitutively expressed along with the profiling of nucleotide sugars UDP-galactose, UDP-arabinose, UDP-glucuronic acid, and UDP-rhamnose in Mara des Bois strawberries under different storage conditions. We also assessed the expression timing of AGP-encoding genes (FvAFP4, FvAGP5) and genes involved in key steps of post-translational glycosylation (FvP4H1, FvGAT20, FvGAT7). Whereas extensins are down-regulated, AGPs are transcriptionally regulated by cold and cold-high CO₂ and post-translationally modulated after transfer to 20 °C. Based on their subcellular localization, molecular properties, isoform-specific glycosylation, UDP-sugar availability, and timing-regulated expression, AGPs are likely involved in cell wall assembly and modulation of mechanical properties. Consequently, they may influence fruit texture and enhanced softening resistance, potentially counteracting senescence-associated disorders through CO₂-responsive signaling mechanisms. Conversely, the decrease in both UDP-galactose levels and AGPs gene expression in non-cold-stored senescent strawberries at 20 °C further supports their relevance in AGPs biosynthesis regulation and underscores their potential as markers for improving postharvest storage strategies. Full article

Background/Objectives: In Ecuador, the prevalence of type 2 diabetes mellitus (T2DM) is the second leading cause of death after ischemic heart disease. Genetic variability in protein-coding genes, single nucleotide variants (SNVs), influences the response to antidiabetic drugs. The frequency of SNVs varies among different populations, so studying the ancestral proportions among SNVs is important for personalized medicine in the treatment of T2DM. This study aimed to evaluate the distribution of Native American, European, and African (NATAM, EUR, and AFR) ancestry in 23 allelic variants of the seven genes that encode the relevant enzymes that metabolize antidiabetic drugs in an Ecuadorian population. Methods: Twenty-three allelic variants of seven genes were analyzed in 297 patients with T2DM from Ecuador, and the molecular ancestry of the samples was analyzed considering three ancestral groups, NATAM, EUR, and AFR using 90 ancestry informative markers (AIMs). Allele and ancestry distributions were analyzed using Spearman’s correlation. Results: The Ecuadorian population presents NATAM (61.33%), EUR (34.48%), and AFR (2.60%) ancestry components. CYP2C8*1 and CYP2C9*1 were positively related to NATAM ancestry, while CYP2C8*4 and CYP2C9*2 were positively related to EUR ancestry. CYP2C19*17 was positively correlated to AFR ancestry. The correlation of SLC22A1 variants such as A in rs594709 was positively correlated with NATAM, while GAT in rs72552763 was positive for EUR. The G variant of rs628031 of the SLC22A1 gene was positively correlated with NATAM and negatively correlated with EUR. The C variant of rs2076828 of the SLC22A3 gene was positively correlated with NATAM ancestry. Conclusions: In the Ecuadorian population, a predominance of Native American ancestry has been observed. Among the allelic variants related to enzymes that metabolize antidiabetic drugs, a relationship has been observed between this ancestral component and variants of the CYP2C8*1, CYP2C9*1, SLC22A1 (rs594709 and rs628031), and SLC22A3 (rs2076828) genes. This information is fundamental for the development of strategies for the implementation of personalized medicine programs for Latin American patients. Full article

Ganoderic acid (GA) is a key bioactive component with pharmacological properties that is found in Ganoderma lingzhi, a renowned medicinal mushroom. Currently, the regulatory mechanisms underlying GA biosynthesis in G. lingzhi remain to be further elucidated. In this study, blue light induction was found to significantly enhance the GA content in G. lingzhi. To explore the regulatory mechanism of GA biosynthesis in response to blue light, the blue light receptor WC-2 was identified, and its regulatory role was characterized. The deletion of wc-2 resulted in a significant reduction in both GA content and the accumulation of intermediates compared to the wild-type control strain, largely due to the strong downregulation of key GA biosynthetic genes. Additionally, decreased asexual spore production and reduced expression of sporulation-specific genes were observed with the deletion of wc-2. The overexpression of wc-2 led to greatly enhanced GA accumulation. Under blue light induction, the maximum contents of GA-Mk, GA-T, GA-S, and GA-Me were 2.27-, 2.51-, 2.49-, and 2.08-fold higher, respectively, compared to the control kept in darkness. These results demonstrate that the blue light receptor WC-2 functions as a positive regulator of GA biosynthesis in G. lingzhi, influencing the expression of genes involved in GA biosynthesis and asexual spore production, thereby advancing our understanding of the intricate regulatory network of GA biosynthesis. Full article

Question answering from visually rich documents (VRDs) is the task of retrieving the correct answer to a natural language question by considering the content of textual and visual elements in the document, as well as the pages’ layout. To answer closed-ended questions that require a deep understanding of the hierarchical relationships between the elements, i.e., the full document-level understanding (FDU) task, state-of-the-art graph-based approaches to FDU model the pairwise element relationships in a graph model. Although they incorporate logical links (e.g., a caption refers to a figure) and spatial ones (e.g., a caption is placed below the figure), they currently disregard the semantic similarity among multimodal document elements, thus potentially yielding suboptimal scoring of the elements’ relevance to the input question. In this paper, we propose GRAS-FDU, a new graph attention network tailored to FDU. GATS-FDU is trained to jointly consider multiple document facets, i.e., the local, spatial, and semantic elements’ relationships. The results show that our approach achieves superior performance compared to several baseline methods. Full article

The proliferation of fake profiles on social media presents critical cybersecurity and misinformation challenges, necessitating robust and scalable detection mechanisms. Such profiles weaken consumer trust, reduce user engagement, and ultimately harm brand reputation and platform credibility. As adversarial tactics and synthetic identity generation evolve, traditional rule-based and machine learning approaches struggle to detect evolving and deceptive behavioral patterns embedded in dynamic user-generated content. This study aims to develop an AI-driven, multi-modal deep learning-based detection system for identifying fake profiles that fuses textual, visual, and social network features to enhance detection accuracy. It also seeks to ensure scalability, adversarial robustness, and real-time threat detection capabilities suitable for practical deployment in industrial cybersecurity environments. To achieve these objectives, the current study proposes an integrated AI system that combines the Robustly Optimized BERT Pretraining Approach (RoBERTa) for deep semantic textual analysis, ConvNeXt for high-resolution profile image verification, and Heterogeneous Graph Attention Networks (Hetero-GAT) for modeling complex social interactions. The extracted features from all three modalities are fused through an attention-based late fusion strategy, enhancing interpretability, robustness, and cross-modal learning. Experimental evaluations on large-scale social media datasets demonstrate that the proposed RoBERTa-ConvNeXt-HeteroGAT model significantly outperforms baseline models, including Support Vector Machine (SVM), Random Forest, and Long Short-Term Memory (LSTM). Performance achieves 98.9% accuracy, 98.4% precision, and a 98.6% F1-score, with a per-profile speed of 15.7 milliseconds, enabling real-time applicability. Moreover, the model proves to be resilient against various types of attacks on text, images, and network activity. This study advances the application of AI in cybersecurity by introducing a highly interpretable, multi-modal detection system that strengthens digital trust, supports identity verification, and enhances the security of social media platforms. This alignment of technical robustness with brand trust highlights the system’s value not only in cybersecurity but also in sustaining platform credibility and consumer confidence. This system provides practical value to a wide range of stakeholders, including platform providers, AI researchers, cybersecurity professionals, and public sector regulators, by enabling real-time detection, improving operational efficiency, and safeguarding online ecosystems. Full article

To overcome the limitations of traditional convolutional and recurrent neural networks in capturing spatio-temporal dynamics in flow fields on unstructured grids, this study proposes a novel Spatio-Temporal Graph Neural Network (ST-GNN) model that integrates a Graph Neural Network (GNN) with a Long Short-Term Memory (LSTM) network. The GNN component captures spatial dependencies among irregular grid nodes via message passing, while the LSTM component models temporal evolution through gated memory mechanisms. This hybrid framework enables the joint learning of spatial and temporal features in complex flow systems. Two variants of ST-GNN, namely, GCN-LSTM and GAT-LSTM, were developed and evaluated using the SUBOFF AFF-8 benchmark dataset. The results show that GAT-LSTM achieved higher accuracy than GCN-LSTM, with average relative errors of 2.51% for velocity and 1.43% for pressure at the 1000th time step. Both models achieved substantial speedups over traditional CFD solvers, with GCN-LSTM and GAT-LSTM accelerating predictions by approximately 350 and 150 times, respectively. These findings position ST-GNN as an efficient and accurate alternative for spatio-temporal flow modeling on unstructured grids, advancing data-driven fluid dynamics. Full article