Search Results (25,264)

The manufacturing industry is rapidly moving toward Artificial Intelligence (AI)-driven autonomous manufacturing, which requires distributed Edge AI architectures in which intelligent devices collaborate in real time. However, the practical deployment of Edge AI is hindered by the lack of standardized, asset-centric integration across heterogeneous devices. This study presents an Asset Administration Shell (AAS)-based Edge AI framework that enables interoperable and coordinated operation among Edge devices through standardized digital asset representations and OPC UA-based communication. In the proposed framework, each Edge device is represented as an AAS-compliant digital assets, enabling both direct inter-edge coordination and centralized asset management. To demonstrate the feasibility of the framework, a functional prototype was implemented consisting of a Raspberry Pi-based Vision Inspector, an autonomous mobile robot (AMR), and an AAS-based monitoring server. Vision-based fault detection is performed directly at the Edge and transmitted in real time to the AMR and the AAS Server, enabling event-driven autonomous response and system-level monitoring. Experimental results show that real-time fault detection and response can be achieved on resource-constrained edge devices while maintaining standardized, asset-level information exchange and interoperability across heterogeneous assets. These results indicate that the AAS-based Edge AI framework provides a practical and scalable foundation for asset-centric autonomous manufacturing systems requiring both real-time operational intelligence and systematic asset management. Full article

Social networks are now integral to corporate strategy and daily social life. They enable the rapid and extensive dissemination of information, proving highly effective for promoting hotel marketing content. Consequently, they facilitate interaction and engagement between hotels and their customers, serving both advertising and evaluation purposes. This study aims to analyse the use of the Facebook social network by luxury wine hotels located in countries associated with the Mediterranean Diet. An analytical model examining the variables of content, interactivity, and visibility was employed. A total of 17 luxury hotel pages were analysed, with data collected using the Karma Fanpage platform, an online tool for social media analysis and monitoring. The findings indicate that the majority of profile posts were photographs, and that this format generated the highest number of user reactions. It is recommended that hotels publish more photographic content to foster greater engagement and conduct further analysis of the specific types of posts that elicit the most reactions. Full article

Static honeypot deployment and one-shot attack-path analysis often become ineffective against adaptive adversaries because fixed decoy layouts are easy to fingerprint and risk estimates quickly go stale. This paper presents a unified, mathematically grounded TrapManager framework that couples graph representation learning with budget-constrained combinatorial optimization for dynamic cyber deception. We model attacker progression on vulnerability-based attack graphs and learn context-aware node embeddings using a Graph Attention Network (GAT) that fuses vulnerability-driven risk signals (e.g., CVSS-derived node scores) with structural features. The learned representations are used to estimate edge plausibility and rank candidate source–target routes at the path level. Given limited resources, we formulate pointTrap placement as a Mixed-Integer Programming (MIP) problem that maximizes the expected interception of high-risk paths while penalizing deployment cost under explicit budget constraints, including mandatory coverage of the top-ranked critical paths. To enable online adaptiveness, a pointTrap-triggered, event-driven feedback mechanism locally amplifies risk around alerted regions, updates path weights without retraining the GAT, and re-solves the MIP for rapid redeployment. Experiments on MulVAL-generated benchmark attack graphs and cross-domain transfer settings demonstrate fast convergence, strong discrimination between attack and non-attack edges, and early interception within a small number of hops even with minimal decoy budgets. Overall, the proposed framework provides a scalable and resource-efficient approach to closed-loop attack-path defense by integrating attention-based learning and integer optimization. Full article

We develop a PDE and boundary integral framework for quaternion-valued fields on product domains $\Omega \subset \mathbb{H}\times \mathbb{H}$ governed by the mixed left/right Cauchy–Fueter system We identify the natural compatibility condition and prove local solvability with quantitative $H^1$ estimates, as well as global weak solvability on admissible products $U_x\times U_y$. Motivated by these estimates, we introduce domains of hyperholomorphy and hyper-conjugates for data that are harmonic in each factor (${\Delta }_{x}u={\Delta }_{y}u=0$), and we establish Carleman-type quantitative unique continuation tools (boundary blow-up, three-balls, and doubling), including a propagation-of-smallness principle across the two factors. On the potential-theoretic side, we construct a double boundary integral representation for biregular fields with kernel $K(\xi ,\eta ;x,y)=E(\xi -x)E(y-\eta )$, establish mapping and jump relations for the associated layer potentials on Lipschitz boundaries, and obtain a Fredholm boundary integral equation for the boundary density in the smooth admissible regime. Finally, we prove a constructive Runge approximation theorem on admissible products and outline a practical discretization workflow consistent with the analysis. Full article

Despite having approximately 24 million native speakers, Romanian remains a low-resource language for automatic speech recognition (ASR), with few accurate and publicly available systems. To address this gap, this study explores the challenges of adapting modern speech recognition models, such as wav2vec 2.0 and Conformer, to Romanian. Our investigation is a comprehensive analysis of the two models, their capabilities to adapt to Romanian data, and the performance of the trained models. The research also focuses on unique attributes of the Romanian language, data collection techniques, including weakly supervised learning, and processing methodologies. Building on the previously introduced Echo dataset of 378 h, we release CRoWL (Crawled Romanian Weakly Labeled), a weakly supervised dataset of 9000 h created via automatic transcription. We obtain strong results that, to the best of our knowledge, are competitive with or exceed publicly reported results for Romanian under comparable open evaluation settings, with Conformer attaining 3.01% WER on Echo + CRoWL and wav2vec 2.0 reaching 4.04% (Echo) and 4.17% (Echo + CRoWL). In addition to the datasets, we also release our most capable models as open source, along with their training plans, thereby providing a solid foundation for researchers interested in languages with limited representation. Full article

Spaceborne infrared ship detection holds critical strategic significance in both military and civilian domains. As a crucial data source for ship detection, infrared remote sensing imagery offers the advantages of all-weather detection and strong anti-interference capability. However, existing methods often overlook the detailed features of small ships and fail to effectively suppress interference, leading to missed detections and false alarms in complex backgrounds. To tackle this issue, this study proposes a hierarchical frequency- and spatial-feature attention network (HFS-Net) for fast and accurate ship detection in spaceborne infrared images. The main motivation is to aggregate frequency-spatial information for improved feature extraction, while devising novel hybrid attention-based structures to facilitate interaction among semantic information. Specifically, we design an adaptive frequency-spatial feature attention (AFSA) module to enrich the feature representation. In particular, AFSA integrates information from spatial and frequency domains and introduces channel attention to adaptively extract important features and edge details of ship targets. In addition, we propose an attention-based component-wise feature interaction (ACFI) module that combines multi-head self-attention to capture long-range feature dependencies and component-wise feature aggregation to further enhance the interaction of high-level semantic information. Extensive experiments demonstrate that HFS-Net achieves higher detection accuracy than several representative detectors in maritime infrared scenes with small ships and complex backgrounds, while maintaining real-time efficiency and moderate computational complexity. Full article

Driving style recognition plays a crucial role in improving the operational safety, fuel efficiency, and intelligent control of commercial vehicles. Under real-world driving conditions, Controller Area Network (CAN) bus data from commercial vehicles simultaneously contain rapid transient variations induced by pedal and braking operations, as well as long-term behavioral trends reflecting driving habits, exhibiting pronounced multi-temporal characteristics. In addition, such data are typically affected by high noise levels, high dimensionality, and highly variable operating conditions, which makes it difficult for methods relying on single-scale features or handcrafted rules difficult to maintain robust and stable performance in complex scenarios. To address these challenges, this paper proposes a driving style classification network, termed the Multi-Scale Convolution and Efficient Channel Attention Network (MSCA-Net). By employing parallel convolutional branches with different temporal receptive fields, the proposed network is able to capture fast driver responses, local temporal dependencies, and long-term behavioral evolution, enabling unified modeling of cross-scale temporal patterns in driving behavior. Meanwhile, the Efficient Channel Attention mechanism adaptively emphasizes CAN signal channels that are highly relevant to driving style discrimination, thereby enhancing the discriminative capability and robustness of the learned feature representations. Experiments conducted on real-world multi-dimensional CAN time-series data collected from commercial vehicles demonstrate that the proposed MSCA-Net achieves improved classification performance in driving style recognition. Furthermore, the potential application of the recognized driving styles in adaptive Automated Manual Transmission shift strategy adjustment is discussed, providing a feasible engineering pathway toward behavior-aware intelligent control of commercial vehicle powertrains. Full article

Stance detection has emerged as an essential tool in natural language processing for understanding how individuals express agreement, disagreement, or neutrality toward specific targets in social and online discourse. It plays a crucial role in bilingual and multilingual environments, including English-Arabic social media ecosystems, where differences in language structure, discourse style, and data availability pose significant challenges for reliable stance modelling. Existing approaches often struggle with target awareness, cross-lingual generalization, robustness to noisy user-generated text, and the interpretability of model decisions. This study aims to build a reliable, explainable target-aware bilingual stance-detection framework that generalizes across heterogeneous stance formats and languages without retraining on a dataset specific to the target language. Thus, a unified dual-encoder architecture based on mDeBERTa-v3 is proposed. Cross-language contrastive learning offers an auxiliary training objective to align English and Arabic stance representations in a common semantic space. Robustness-oriented regularization is used to mitigate the effects of informal language, vocabulary variation, and adversarial noise. To promote transparency and trustworthiness, the framework incorporates token-level rationale extraction, enables fine-grained interpretability, and supports analysis of hallucination. The proposed model is tested on a combined bilingual test set and two structurally distinct zero-shot benchmarks: MT-CSD and AraStance. Experimental results show consistent performance, with accuracies of 85.0% and 86.8% and F1-scores of 84.7% and 86.8% on the zero-shot benchmarks, confirming stable performance and realistic generalization. Ultimately, these findings reveal that effective bilingual stance detection can be achieved via explicit target conditioning, cross-lingual alignment, and explainability-driven design. Full article

Speech-driven face generation aims to synthesize a face image that matches a speaker’s identity from speech alone. However, existing methods typically trade identity fidelity for visual quality and rely on large end-to-end generators that are difficult to train and tune. We propose Vox2Face, a speech-driven face generation framework centered on an explicit identity space rather than direct speech-to-image mapping. A pretrained speaker encoder first extracts speech embeddings, which are distilled and metric-aligned to the ArcFace hyperspherical identity space, transforming cross-modal regression into a geometrically interpretable speech-to-identity alignment problem. On this unified identity representation, we reused an identity-conditioned diffusion model as the generative backbone and synthesized diverse, high-resolution faces in the Stable Diffusion latent space. To better exploit this prior, we introduce a discriminator-free diffusion self-consistency loss that treats denoising residuals as an implicit critique of speech-predicted identity embeddings and updates only the speech-to-identity mapping and lightweight LoRA adapters, encouraging speech-derived identities to lie on the high-probability identity manifold of the diffusion model. Experiments on the HQ-VoxCeleb dataset show that Vox2Face improves the ArcFace cosine similarity from 0.295 to 0.322, boosts R@10 retrieval accuracy from 29.8% to 32.1%, and raises the VGGFace Score from 18.82 to 23.21 over a strong diffusion baseline. These results indicate that aligning speech to a unified identity space and reusing a strong identity-conditioned diffusion prior is an effective method to jointly improve identity fidelity and visual quality. Full article

Groundwater, a vital resource for drinking water supply, must be managed sustainably to ensure its availability and quality. In France, the SISE-Eaux database on water intended for human consumption, archived by the Regional Health Agencies (ARS) since 1990, constitutes a rich source of information. This study focused on the groundwater of the Nouvelle-Aquitaine region, the largest administrative region in metropolitan France, covering 84,061 km² with 6 million inhabitants. It is based on a 22-year data extraction, resulting in a matrix of 121,649 observations and 51 physico-chemical and bacteriological parameters. Following logarithmic transformation of the data and fitting of variograms using the mean value of each parameter for each sampling point, the spatial distribution of numerous parameters across the region is presented. From this initial sparse matrix, a dense matrix of 23,319 samples (rows) and 15 key parameters (columns) was selected for a multivariate approach. A Principal Component Analysis (PCA) was used to condense the information and create summary maps capturing over 68% of the information contained in the dense matrix. The combined results of the multivariate analysis (dense matrix) and the distribution of individual parameters (sparse matrix) highlight the diversity of sources contributing to the spatial variability of groundwater, such as the role of lithology, the origin and pathways of fecal contamination, and the influence of redox processes. Neither the large size of the study area nor the high number of parameters proved to be an obstacle to the analysis. The understanding of ongoing processes and the factorial axis distribution maps, which enable the spatial representation of these mechanisms, can be used to facilitate groundwater monitoring and protection. Full article

We aim to address the insufficient robustness of navigational line detection for rapeseed seed production sires in complex field scenarios and the challenges faced by existing models in balancing precision, real-time performance, and resource consumption. Taking YOLOv8n-seg as the baseline, we first introduced the ADown module to mitigate feature subsampling information loss and enhance computational efficiency. Subsequently, the DySample module was employed to strengthen target feature representation and improve object discrimination in complex scenarios. Finally, the c2f module was replaced with c2f_FB to optimise feature fusion and reinforce multi-scale feature integration. Performance was evaluated through comparative experiments, ablation studies, and scenario testing. The model achieves an average precision of 99.2%, mAP50-95 of 84.5%, a frame rate of 90.21 frames per second, and 2.6 million parameters, demonstrating superior segmentation performance in complex scenarios. SegNav-YOLOv8n balances performance and resource requirements, validating the effectiveness of the improvements and providing reliable technical support for navigating agricultural machinery in rapeseed seed production. Full article

The pursuit of reliable mobile robot autonomy continues to hinge on the nuanced configuration of its navigation subsystems. Within the widely adopted Robot Operating System (ROS), the layered costmap serves as the critical environmental representation, yet its numerous parameters are notoriously difficult to tune by hand. This empirical struggle often leads to deployments that are either overly cautious or dangerously optimistic. Our investigation focuses on demystifying this process through a structured analysis of two pivotal parameters, the inflation radius and the robot radius. We conducted a series of simulated navigation trials using the Hiwonder JetAcker platform in ROS Noetic, meticulously measuring outcomes related to path success, efficiency, and adherence to safety margins. These simulation findings were then cautiously validated through a set of targeted hardware experiments. Our results demonstrate that empirically best-performing inflation radius of 0.3 m for the Hiwonder JetAcker platform reduces failure rates in constrained spaces by 40% compared to the conventional default of 0.6 m. Furthermore, our work quantifies the gap between simulated and real-world navigation. The resulting framework offers practitioners a more informed methodology for systematic parameter tuning and underscores the non-trivial consequences of seemingly minor configuration changes. Full article

Biological muscles generate tension from the combined contribution of the passive elastic recoil and the actively controlled contractile mechanisms. Understanding and replicating these passive and active tensions is necessary and beneficial for designing soft robotic actuators that emulate muscle-like behavior. In the current work, the aim is to develop a mathematical framework for modeling both the passive and active tensions in a biological muscle as functions of muscle length and contraction velocity. We will describe the passive tension by a nonlinear monotonically increasing function of length with threshold behavior in order to capture the experimentally observed stiffening occurring in stretched biological muscles. We will model the active tension using the superposition of Gaussian functions that relate bell-shaped tension-length with a flat plateau over the optimal length of the sarcomere. The parameters of this Gaussian representation of the active tension-length relation are determined from formulating a least-squares optimization problem, such that a Characteristic (indicator) function is approximated globally over the optimal length range of the sarcomere by summation of some Gaussian functions. The closed-form formulations for the required integrals are derived using the integral of the product of two Gaussian functions over ${\mathbb{R}}^n$ as well as the error function which enables efficient parameter identification. We will also propose a symmetric tension–velocity relation that distinguishes three phases of concentric, eccentric and isometric contractions, and is parametrized directly by measurable quantities of isometric tension and maximum shortening velocity. The passive and active tensions are finally combined into a unified comprehensive tension model in which the exponentially modeled passive tension is added up to the active contribution, formulated as the product of the activation level, a normalized length-dependent factor and a normalized velocity-dependent factor. The resulting model reproduces canonical tension-length and tension-velocity relations and provides an analytically tractable comprehensive tension model that can be embedded in the dynamics of soft and continuum robot actuators inspired by biological muscles. Full article

Node classification is a fundamental task in graph-based learning, with applications in social networks, citation networks, and biological systems. Learning node representations for different graph datasets is necessary to find the correlation between different types of nodes. Graph Neural Networks (GNNs) play a critical role in providing revolutionary solutions for graph data structures. In this paper, we analyze the effect of combined GNN and multilayer perceptron (MLP) architecture to solve the node classification problem for different graph datasets. The feature information and network topology are efficiently captured by the GNN layer, and the MLP helps to make accurate decisions. We have selected popular datasets, namely Amazon-computer, Amazon-photo, Citeseer, Cora, Corafull, PubMed, and Wikics, for evaluating the performance of the proposed approach. In addition, in the GNN part, we have used six models to find the best model fit in the proposed architecture. We have conducted extensive simulations to find the node classification accuracy for the proposed model. The results show the proposed architecture can outperform previous studies in terms of test accuracy. In particular, the GNN algorithms SAGEConv, GENConv, and TAGConv show superior performance across different datasets. Full article

Cross-domain sentiment analysis remains challenging due to distributional shifts and heterogeneous sentiment expressions across platforms. Existing domain adaptation approaches primarily focus on enforcing domain-invariant representations. However, such symmetry-preserving strategies often overlook directional and expression-level asymmetries. These asymmetries naturally arise in real-world sentiment data, particularly for context-inferred sentiment expressions. In this work, we propose a novel symmetry- and asymmetry-aware domain adaptation framework for cross-domain sentiment classification. The framework models symmetry through explicit multi-source distribution alignment, which captures transferable sentiment knowledge across domains. Additionally, aspect-level structural supervision organizes representations according to shared linguistic aspects. To address asymmetry, a directional divergence regularization is introduced. This component models expression-level and directional discrepancies between source and target domains. Importantly, the framework operates without requiring target-domain annotations. Experiments are conducted under a multi-source unsupervised domain adaptation setting using sentence-level hotel review datasets collected from multiple online platforms. Empirical results demonstrate strong performance for the proposed framework. It achieves an average Accuracy of 82.0% and Macro-F1 of 80.6%. The framework consistently and statistically significantly outperforms source-only, multi-source, and transformer-based adversarial adaptation baselines across all evaluated target domains (p < 0.05). Additional analyses confirm improved robustness to implicit sentiment expressions and platform-induced asymmetries. These findings highlight the importance of jointly modeling symmetry and asymmetry for robust cross-domain sentiment adaptation and provide a unified and deployable solution for sentiment analysis under realistic platform shifts. Full article