Search Results (6,432)

Blended learning is widely adopted in higher education, yet little is known about how students experience its motivational and instructional features. In this study, we examined undergraduate students’ experiences regarding blended learning by integrating Herzberg’s two-factor theory with the TPACK framework. Semi-structured interviews were conducted with 24 undergraduates at a large Vietnamese university. A theory-informed qualitative content analysis approach was used to identify codes, categories, and themes. These were then mapped onto the pedagogical content knowledge (PCK), technological content knowledge (TCK), and technological pedagogical knowledge (TPK) intersections of the TPACK framework. The findings showed that hygiene factors included unengaging teaching practices, inadequate digital infrastructure, and limited online interaction. These factors often produced frustration and reduced engagement. Motivator factors included active and relevant pedagogical strategies, engaging and accessible digital resources, and technology-facilitated autonomous, expressive, and creative learning work. These factors encouraged deeper learning and stronger motivation. It is concluded that blended learning design must address both hygiene and motivator factors to improve student engagement. Integrating these factors with the TPACK intersections offers a practical model for improved course structures, enhanced digital resources, and the design of more interactive technology-supported pedagogy. The findings provide actionable implications for higher education institutions seeking to improve the quality of blended learning. Full article

This paper foregrounds the study life of students in Early Childhood and Social Education through the concept of educational belonging, conceptualized as situated, relational, affective, and spatial practices that are continually renegotiated. As an affective and spatial practice, educational belonging foregrounds that places, spatial designs, and interiors play a constitutive role in shaping study life, including students’ study experiences and learning processes. The paper is based on a research project conducted at University College Copenhagen, which investigates the significance of educational architecture for students’ learning processes and sense of belonging within their education. Drawing on a substantial body of data generated through architectural plan interviews, the research project offers insight into how the design and atmosphere of educational spaces and places co-constitute students’ sensory experiences of belonging. The analytical parts of the paper illuminate how students’ narratives about their positioning within classrooms reveal that teaching and learning situations are social and affective events, in which students are recognized as occupying specific student positions, such as serious, nerdy, or disengaged. Full article

Offensive-language detection systems that perform well at a given point in time often degrade as linguistic patterns evolve, particularly in dialectal Arabic social media, where new terms emerge and familiar expressions shift in meaning. This study investigates temporal linguistic drift in Saudi-dialect offensive-language detection through a systematic evaluation of continual-learning approaches. Building on the Saudi Offensive Dialect (SOD) dataset, we designed test scenarios incorporating newly introduced offensive terms, context-shifting expressions, and varying proportions of historical data to assess both adaptation and knowledge retention. Eight continual-learning configurations—Experience Replay (ER), Elastic Weight Consolidation (EWC), Low-Rank Adaptation (LoRA), and their combinations—were evaluated across five test scenarios. Results show that models without continual-learning experience a 13.4-percentage-point decline in F1-macro on evolved patterns. In our experiments, Experience Replay achieved a relatively favorable balance, maintaining 0.812 F1-macro on historical data and 0.976 on contemporary patterns (KR = −0.035; AG = +0.264), though with increased memory and training time. EWC showed moderate retention (KR = −0.052) with comparable adaptation (AG = +0.255). On the SimuReal test set—designed with realistic class imbalance and only 5% drift terms—ER achieved 0.842 and EWC achieved 0.833, compared to the original model’s 0.817, representing modest improvements under realistic conditions. LoRA-based methods showed lower adaptation in our experiments, likely reflecting the specific LoRA configuration used in this study. Further investigation with alternative settings is warranted. Full article

Entity alignment (EA) facilitates knowledge fusion by matching semantically identical entities in distinct knowledge graphs (KGs). Existing embedding-based methods rely solely on intrinsic KG facts and often struggle with long-tail entities due to insufficient information. Recently, large language models (LLMs), empowered by rich background knowledge and strong reasoning abilities, have shown promise for EA. However, most current LLM-enhanced approaches follow the in-context learning paradigm, requiring multi-round interactions with carefully designed prompts to perform additional auxiliary operations, which leads to substantial computational overhead. Moreover, they fail to fully exploit the complementary strengths of embedding-based small models and LLMs. To address these limitations, we propose HIEA, a novel hierarchical inference framework for entity alignment. By instruction-tuning a generative LLM with a unified and concise prompt and a knowledge adapter, HIEA produces alignment results with a single LLM invocation. Meanwhile, embedding-based small models not only generate candidate entities but also support the LLM through data augmentation and certainty-aware source entity classification, fostering deeper collaboration between small models and LLMs. Extensive experiments on both standard and highly heterogeneous benchmarks demonstrate that HIEA consistently outperforms existing embedding-based and LLM-enhanced methods, achieving absolute Hits@1 improvements of up to 5.6%, while significantly reducing inference cost. Full article

The increasing size of recent deep neural networks, particularly when applied to learning across multiple domains, limits their deployment in resource-constrained environments. To address this issue, this study proposes a lightweight neural architecture with a parallel structure of convolutional layers to enable efficient and scalable multi-domain learning. The proposed network includes an individual feature extractor for domain-specific features and a common feature extractor for the shared features. This design minimizes redundancy and significantly reduces the number of parameters while preserving classification performance. To evaluate the proposed method, experiments were conducted using four image classification datasets: MNIST, FMNIST, CIFAR10, and SVHN. These experiments focused on classification settings where each image contained a single dominant object without relying on large pretrained models. The proposed model achieved high accuracy while significantly reducing the number of parameters. It required only 3.9 M parameters for learning across the four datasets, compared to 33.6 M for VGG16. The model achieved an accuracy of 98.87% on MNIST and 85.83% on SVHN, outperforming other lightweight models, including MobileNet v2 and EfficientNet v2b0, and was comparable to ResNet50. These findings indicate that the proposed architecture has the potential to support multi-domain learning while minimizing model complexity. This approach may be beneficial for applications in resource-constrained environments. Full article

Videogames (VGs) are highly attractive for children and young people. Although videogames were once viewed mainly as sources of distraction and leisure, they are now widely recognised as powerful tools for competence development across diverse domains. Designing and implementing a videogame is even more appealing for children and novice students than merely playing it, but developing programming competencies using a text-based language often constitutes a significant barrier to entry. This article presents the implementation and evaluation of a videogame development experience with university students using the Unity engine and its Visual Scripting block-based tool. Students worked in teams and successfully completed videogame projects, demonstrating substantial gains in programming and game construction skills. The adopted methodology facilitated learning, collaboration, and engagement. Building on a quasi-experimental design that compared a prior unit based on C# and MonoGame with a subsequent unit based on Unity Visual Scripting, the study analyses differences in performance, development effort, and motivational indicators. The results show statistically significant improvements in grades, reduced development time for core mechanics, and higher self-reported confidence when Visual Scripting is employed. The evidence supports the view of Visual Scripting as an effective educational strategy to introduce programming concepts without the syntactic and semantic barriers of traditional text-based languages. The findings further suggest that Unity Visual Scripting can act as a didactic bridge towards advanced programming, and that its adoption in secondary and primary education is promising both for reinforcing traditional subjects (history, language, mathematics) and for fostering foundational programming and videogame development skills in an inclusive manner. Full article

Accurate traffic flow prediction is a core component of intelligent transportation systems, supporting proactive traffic management, resource optimization, and sustainable urban mobility. However, urban traffic networks exhibit heterogeneous multi-scale periodic patterns and time-varying spatial interactions among road segments, which are not sufficiently captured by many existing spatio-temporal forecasting models. To address this limitation, this paper proposes PDR-STGCN (Periodicity-Aware Dynamic Relational Spatio-Temporal Graph Convolutional Network), an enhanced STGCN framework that jointly models multi-scale periodicity and dynamically evolving spatial dependencies for traffic flow prediction. Specifically, a periodicity-aware embedding module is designed to capture heterogeneous temporal cycles (e.g., daily and weekly patterns) and emphasize dominant social rhythms in traffic systems. In addition, a dynamic relational graph construction module adaptively learns time-varying spatial interactions among road nodes, enabling the model to reflect evolving traffic states. Spatio-temporal feature fusion and prediction are achieved through an attention-based Bidirectional Long Short-Term Memory (BiLSTM) network integrated with graph convolution operations. Extensive experiments are conducted on three datasets, including Metro Traffic Los Angeles (METR-LA), Performance Measurement System Bay Area (PEMS-BAY), and a real-world traffic dataset from Guizhou, China. Experimental results demonstrate that PDR-STGCN consistently outperforms state-of-the-art baseline models. For next-hour traffic forecasting, the proposed model achieves average reductions of 16.50% in RMSE, 9.00% in MAE, and 0.34% in MAPE compared with the second-best baseline. Beyond improved prediction accuracy, PDR-STGCN reveals latent spatio-temporal evolution patterns and dynamic interaction mechanisms, providing interpretable insights for traffic system analysis, simulation, and AI-driven decision-making in urban transportation networks. Full article

The growing global reliance on solar photovoltaic (PV) systems requires robust, automated inspection techniques to ensure reliability and efficiency. Thermal infrared (IR) imaging is widely used for detecting PV faults; however, accurate classification remains challenging due to severe class imbalance, low thermal contrast, and high inter-class visual similarity among fault types. This study proposes a hierarchical deep learning framework for thermal PV fault classification, integrating a multi-class dataset-balancing strategy to enhance representational efficiency. The proposed framework consists of two major components: (i) a hierarchical two-stage classification scheme that mitigates data imbalance and leverages limited labeled data for improved fault discrimination; and (ii) a contrast-preserving MixUp augmentation technique designed explicitly for low-contrast thermal imagery, improving minority fault class recognition and overall robustness. Comprehensive experiments were conducted on benchmark 8-class thermal PV datasets using nine deep network architectures. Dataset refactoring decisions are validated through quantitative inter-class distance analysis using multiple complementary metrics. Results demonstrate that the proposed hierarchical SlantNet model achieves the best trade-off between accuracy and computational efficiency, achieving an F1-Efficiency Index of 337.6 and processing 42,072 images per second on a GPU, over twice the efficiency of conventional approaches. Comparatively, the Swin-T Transformer attained the highest classification accuracy of 89.48% and F1 score of 80.50%, while SlantNet achieved 86.15% accuracy and 73.03% F1 score with substantially higher inference speed, highlighting its real-time potential. Ablation studies on augmentation and regularization strategies confirm that the proposed techniques significantly improve minority class detection without compromising overall performance, with detailed per-class precision, recall, and F1 analysis. The proposed framework delivers a high-accuracy, low-latency, and edge-deployable solution for automated PV inspection, facilitating seamless integration into operational PV plants for real-time fault diagnosis. Full article

AI technologies are increasingly applied to optimize operations in both port and urban logistics systems, yet integration across the full maritime city chain remains limited. The objective of this study is to assess, using a simulation-based experiment, the impact of an AI-orchestrated control policy on the performance of port–city logistics relative to a baseline scheduler. The study proposes an AI-orchestrated approach that connects autonomous ships, smart ports, central warehouses, and multimodal urban networks via a shared cloud control layer. This approach is designed to enable real-time, cross-domain coordination using federated sensing and adaptive control policies. To evaluate its impact, a simulation-based experiment was conducted comparing a traditional scheduler with an AI-orchestrated policy across 20 paired runs under identical conditions. The orchestrator dynamically coordinated container dispatching, vehicle assignment, and gate operations based on capacity-aware logic. Results show that the AI policy substantially reduced the total completion time, lowered truck idle time and estimated emissions, and improved system throughput and predictability without modifying physical resources. These findings support the expectation that integrated, data-driven decision-making can significantly enhance logistics performance and sustainability in port–city contexts. The study provides a replicable pathway from conceptual architecture to quantifiable evidence and lays the groundwork for future extensions involving learning controllers, richer environmental modeling, and real-world deployment in digitally connected logistics corridors. Full article

High-Resolution Range Profile (HRRP) represents the electromagnetic backscattering distribution of targets and plays a pivotal role in remote-sensing-based Automatic Target Recognition (RATR). However, in non-cooperative sensing scenarios, acquiring sufficient measured data is severely constrained by operational costs and physical limitations, leading to data scarcity that hampers model robustness. To overcome this, we propose SpecM-DDPM, a spectral-aware Denoising Diffusion Probabilistic Models (DDPM) tailored for generating high-fidelity HRRPs that preserve physical scattering properties. Unlike generic generative models, SpecM-DDPM incorporates radar signal physics into the diffusion process. Specifically, a parallel multi-scale block is designed to adaptively capture both local scattering centers and global target resonance structures. To ensure spectral fidelity, a spectral gating mechanism serves as a physics-constrained filter to calibrate the energy distribution in the frequency domain. Furthermore, a Frequency-Aware Curriculum Learning (FACL) strategy is introduced to guide the progressive reconstruction from low-frequency structural components to high-frequency scattering details. Experiments on measured aircraft data demonstrate that SpecM-DDPM generates samples with high physical consistency, significantly enhancing the generalization performance of radar recognition systems in data-limited environments. Full article

Digital inclusion has become an essential component in ensuring the autonomy, social participation, and well-being of older adults. However, their learning of digital skills depends to a large extent on the quality of support provided by the facilitator, whose age, training, and experience directly influence teaching processes and how older adults relate to technology. This study compares the digital competences, and ICT skills of 107 facilitators of digital literacy programs, classified into three groups: peer educators (PEERS), young students without gerontological training (YOS), and young gerontology specialists (YGS). A quantitative design was used. Statistical analyses included non-parametric tests (Kruskal–Wallis, Mann–Whitney, Kendall’s Tau) and parametric tests (ANOVA, t-tests), to examine associations between socio-demographic variables, the level of digital competence, and ICT skills for teachers (technological and pedagogical). The results show clear differences between profiles. YOS achieved the highest scores in digital competence, especially in problem-solving and tool handling. The YGS achieved a balanced profile, combining competent levels of digital skills with pedagogical strengths linked to their gerontological training. In contrast, PEERS recorded the lowest levels of digital competence, particularly in security and information management; nevertheless, their role remains relevant for fostering trust and closeness in training processes among people of the same age. It was also found that educational level is positively associated with digital competence in all three profiles, while age showed a negative relationship only among PEERS. The findings highlight the importance of creating targeted training courses focusing on digital, technological, and pedagogical skills to ensure effective, tailored teaching methods for older adults. Full article

Periocular recognition is essential when full-face images cannot be used because of occlusion, privacy constraints, or sensor limitations, yet in many deployments, only periocular images are available at run time, while richer evidence, such as archival face photos and textual metadata, exists offline. This mismatch makes it hard to deploy conventional multimodal fusion. This motivates the notion of conditional biometrics, where auxiliary modalities are used only during training to learn stronger periocular representations while keeping deployment strictly periocular-only. In this paper, we propose Multi-Anchor Conditional Periocular Embedding (MACPE), which maps periocular, facial, and textual features into a shared anchor-conditioned space via a learnable anchor bank that preserves periocular micro-textures while aligning higher-level semantics. Training combines identity classification losses on periocular and face branches with a symmetric InfoNCE loss over anchors and a pulling regularizer that jointly aligns periocular, facial, and textual embeddings without collapsing into face-dominated solutions; captions generated by a vision language model provide complementary semantic supervision. At deployment, only the periocular encoder is used. Experiments across five periocular datasets show that MACPE consistently improves Rank-1 identification and reduces EER at a fixed FAR compared with periocular-only baselines and alternative conditioning methods. Ablation studies verify the contributions of anchor-conditioned embeddings, textual supervision, and the proposed loss design. Full article

Pragmatic language is a core component of school-based social participation, yet children with Autism Spectrum Disorder (ASD) and Social Communication Disorder (SCD) frequently experience persistent difficulties in using language appropriately across everyday learning contexts. This study investigated the effectiveness of a culturally adapted, school-based immersive Virtual Reality (VR) learning program designed to enhance pragmatic language and social communication skills among Thai primary school children. Eleven participants aged 7–12 years completed a three-week, ten-session VR program that simulated authentic classroom, playground, and canteen interactions aligned with Thai sociocultural norms. Outcomes were measured using the Social Communication Questionnaire (SCQ) and the Pragmatic Behavior Observation Checklist (PBOC). While SCQ scores showed a small, non-significant reduction (p = 0.092), PBOC results demonstrated significant improvements in three foundational pragmatic domains: Initiation and Responsiveness (p = 0.032), Turn-Taking and Conversational Flow (p = 0.037), and Politeness and Register (p = 0.010). Other domains showed no significant changes. These findings suggest that immersive, culturally relevant VR environments can support early gains in core pragmatic language behaviors within educational settings, although broader social communication outcomes may require longer or more intensive learning experiences. Full article

Software-defined (SD) honeypots, as dynamic cybersecurity technologies, enhance defense efficiency through flexible resource allocation. However, traditional SD honeypots face latency and jitter issues under network fluctuations, while balancing adjustment costs with defense benefits remains challenging. This paper proposes a DPU-accelerated SD honeypot security service deployment method, leveraging DPU hardware acceleration to optimize network traffic processing and protocol parsing, thereby significantly improving honeypot environment construction efficiency and response real-time performance. For dynamic attack–defense scenarios, we design an adaptive adjustment strategy combining Stackelberg game theory with deep reinforcement learning (AASGRL). By calculating the expected defense benefits and adjustment costs of optimal honeypot deployment strategies, the approach dynamically determines the timing and scope of honeypot adjustments. Simulation experiments demonstrate that the mechanism requires no adjustments in 80% of interaction rounds, while achieving enhanced defense benefits in 20% of rounds with controlled adjustment costs. Compared to traditional methods, the AASGRL mechanism maintains stable defense benefits in long-term interactions, verifying its effectiveness in balancing low costs and high benefits against dynamic attacks. This work provides critical technical support for building adaptive proactive network defense systems. Full article

Reliable and low-latency seizure detection from electroencephalography (EEG) is critical for continuous clinical monitoring and emerging wearable health technologies. Spiking neural networks (SNNs) provide an event-driven computational paradigm that is well suited to real-time signal processing, yet achieving competitive seizure detection performance with constrained model complexity remains challenging. This work introduces a hybrid spike encoding scheme that combines Delta–Sigma (change-based) and stochastic rate representations, together with two spiking architectures designed for real-time EEG analysis: a compact feed-forward HybridSNN and a convolution-enhanced ConvSNN incorporating depthwise-separable convolutions and temporal self-attention. The architectures are intentionally designed to operate on short EEG segments and to balance detection performance with computational practicality for continuous inference. Experiments on the CHB–MIT dataset show that the HybridSNN attains 91.8% accuracy with an F1-score of 0.834 for seizure detection, while the ConvSNN further improves detection performance to 94.7% accuracy and an F1-score of 0.893. Event-level evaluation on continuous EEG recordings yields false-alarm rates of 0.82 and 0.62 per day for the HybridSNN and ConvSNN, respectively. Both models exhibit inference latencies of approximately 1.2 ms per 0.5 s window on standard CPU hardware, supporting continuous real-time operation. These results demonstrate that hybrid spike encoding enables spiking architectures with controlled complexity to achieve seizure detection performance comparable to larger deep learning models reported in the literature, while maintaining low latency and suitability for real-time clinical and wearable EEG monitoring. Full article