Search Results (1,271)

Effective signaling in instructional materials—through cues such as highlights, arrows, and annotations—can guide learner attention, reduce cognitive load, and enhance comprehension in multimedia-rich online courses. While the benefits of signaling are well documented, little is known about how combinations of signaling strategies influence both the average performance and the consistency of student outcomes. In this study, we propose a data-driven approach to evaluate and optimize signaling strategies in online teaching. Using lecture materials from three semesters of introductory and intermediate statistics courses, we extracted multiple features of textual and visual signaling, including highlighted words, annotated formulas, arrows, and notes. Principal Component Analysis identified four distinct signaling strategies employed by the instructor. We then applied a heteroscedastic beta regression model to link these strategies to topic-level exam performance, allowing simultaneous assessment of mean learning outcomes and their variability. Results show that strategies combining formula highlighting with arrows and detailed notes improve both the average proportion of successful learners and the stability of outcomes, while relying solely on formula highlighting increases variability. Our findings provide actionable guidance for instructors to design effective signaling strategies, and demonstrate a flexible framework for data-driven evaluation of teaching practices in online learning environments. Full article

Map matching is a fundamental technique for aligning noisy GPS trajectory data with digital road networks and constitutes a key component of Intelligent Transportation Systems (ITS) and Location-Based Services (LBS). Nevertheless, existing approaches still suffer from notable limitations in complex environments, particularly urban and urban-like scenarios characterized by heterogeneous GPS noise and sparse observations, including inadequate adaptability to dynamically varying noise, unavoidable trade-offs between real-time efficiency and matching accuracy, and limited generalization capability across heterogeneous driving behaviors. To overcome these challenges, this paper presents a Meta-learning-driven Progressive map-Matching (MPM) method with a symmetry-aware design, which integrates a two-layer pattern-mining-based noise-robust meta-learning mechanism with a dynamic weight adjustment strategy. By explicitly modeling topological symmetry in road networks, symmetric trajectory patterns, and symmetric noise variation characteristics, the proposed method effectively enhances prior knowledge utilization, accelerates online adaptation, and achieves a more favorable balance between accuracy and computational efficiency. Extensive experiments on two real-world datasets demonstrate that MPM consistently outperforms state-of-the-art methods, achieving up to 10–15% improvement in matching accuracy while reducing online matching latency by over 30% in complex urban environments. Furthermore, the symmetry-aware design significantly improves robustness against asymmetric interference, thereby providing a reliable and scalable solution for high-precision map matching in complex and dynamic traffic environments. Full article

Hierarchical federated learning (HFL) has emerged as a promising paradigm for distributed machine learning over vehicular networks. Despite recent advances in vehicle selection and resource allocation, most still adopt a fixed Edge-and-Vehicle Scheduling (EVS) configuration that keeps the number of participating edge nodes and vehicles per node constant across training rounds. However, given the diverse training tasks and dynamic vehicular environments, our experiments confirm that such static configurations struggle to efficiently meet the task-specific requirements across model accuracy, time delay, and energy consumption. To address this, we first formulate a unified, long-term training cost metric that balances these conflicting objectives. We then propose AptEVS, an adaptive scheduling framework based on deep reinforcement learning (DRL), designed to minimize this cost. The core of AptEVS is its phase-aware design, which adapts the scheduling strategy by first identifying the current training phase and then switching to specialized strategies accordingly. Extensive simulations demonstrate that AptEVS learns an effective scheduling policy online from scratch, consistently outperforming baselines and and reducing the long-term training cost by up to 66.0%. Our findings demonstrate that phase-aware DRL is both feasible and highly effective for resource scheduling over complex vehicular networks. Full article

Artificial intelligence is reshaping approaches to architectural heritage conservation by enabling a deeper understanding of how people perceive and experience historic built environments. This study employs deep learning and large language models (LLMs) to explore public perceptions of the Qinghefang Historical and Cultural District in Hangzhou, illustrating how AI-driven analytics can inform intelligent heritage management and architectural revitalization. Large-scale public online reviews were processed through BERTopic-based clustering to extract thematic structures of experience, while interpretive synthesis was refined using an LLM to identify core perceptual dimensions including Hangzhou Housing & Residential Choice, Hangzhou Urban Tourism & Culture, Hangzhou Food & Dining, and Qinghefang Culture & Creative. Sentiment polarity and emotional intensity were quantified using a fine-tuned BERT model, revealing distinct affective and perceptual patterns across the district’s architectural and cultural spaces. The results demonstrate that AI-based textual analytics can effectively decode human–heritage interactions, offering actionable insights for data-informed conservation, visitors’ experience optimization, and sustainable management of historic districts. This research contributes to the emerging field of AI-driven innovation in architectural heritage by bridging computational intelligence and heritage conservation practice. Full article

The transition from high school to university-level mathematics is often accompanied by significant challenges. During the COVID-19 pandemic, these difficulties were further exacerbated by the abrupt shift to online learning. In response, educators increasingly turned to gamification—“a process of enhancing a service with affordances for gameful experiences in order to support users’ overall value creation”—as a strategy to address the limitations of remote instruction. In this study, we designed a gamified environment for a first-year Number Theory course. The system was constructed using targeted game elements such as leaderboards, optional challenge exams, and recognition for elegant solutions. These features were then integrated into a comprehensive point-based assessment system, which accounted for weekly quizzes and active participation. Following a quasi-experimental design, this study compared two groups of pre-service mathematics teachers: the class of 2017 (N = 62), which received traditional in-person instruction (control group), and the class of 2020 (N = 61), which participated in an online, gamified version of the course (experimental group). Both groups were taught by the same lecturer, using identical content, concepts, and similar tasks throughout the course. Academic performance was measured using midterm exam results. While no significant difference emerged on the first midterm in week 6 (their average percentages were 50% and 51%), the experimental group significantly outperformed the control group on the second midterm at the end of the term (their average percentages were 65% and 49%). These results suggest that a thoughtfully designed, gamified approach can enhance learning outcomes in an online mathematics course. Full article

Accurate prediction of marine visibility is critical for ensuring safe and efficient maritime operations, particularly in dynamic and data-sparse ocean environments. Although visibility reduction is a natural and unavoidable atmospheric phenomenon, improved short-term prediction can substantially enhance navigational safety and operational planning. While deep learning methods have demonstrated strong performance in land-based visibility prediction, their effectiveness in marine environments remains constrained by the lack of fixed observation stations, rapidly changing meteorological conditions, and pronounced spatiotemporal variability. This paper introduces SeADL, a self-adaptive deep learning framework for real-time marine visibility forecasting using multi-source time-series data from onboard sensors and drone-borne atmospheric measurements. SeADL incorporates a continuous online learning mechanism that updates model parameters in real time, enabling robust adaptation to both short-term weather fluctuations and long-term environmental trends. Case studies, including a realistic storm simulation, demonstrate that SeADL achieves high prediction accuracy and maintains robust performance under diverse and extreme conditions. These results highlight the potential of combining self-adaptive deep learning with real-time sensor streams to enhance marine situational awareness and improve operational safety in dynamic ocean environments. Full article

Informal learning increasingly occurs in digital environments, where citizens access, evaluate and apply knowledge outside of formal education. In the context of disaster preparedness, such informal learning is crucial for promoting individual and collective self-protection. This study examines how disaster preparedness knowledge is represented in German-language online resources, and how these materials can be categorised from an adult education perspective. An exploratory mixed-methods design combining expert-guided sampling, a qualitatively developed coding scheme, large-scale web scraping and AI-assisted classification was employed. A total of 7305 webpages were analysed in terms of actor type, topic, media format, and didactic design. The findings suggest that government and commercial organisations dominate the online preparedness landscape, with limited contributions from civil society and individuals. Thematically, most resources focus on general preventive measures and checklists, whereas scenario-specific and procedural content is underrepresented. Didactically rich and interactive formats are rare, with most materials relying on static, text-based communication. From an adult education perspective, these results suggest a gap between raising awareness and active learning. While online resources offer easy access to preparedness knowledge, they rarely facilitate deeper understanding, participation or collaborative learning. Methodologically, the study illustrates how AI-assisted analysis can combine qualitative interpretive depth with computational scalability in educational research. Full article

The Erasmus+ YOU4BLUE project represents an interdisciplinary educational initiative aimed at fostering a youth culture of sustainability through hands-on learning, scientific literacy, and critical thinking focused on the marine environment. The project aimed to encourage lasting behavioural change and empower young people to act. It engaged secondary school students aged 14 to 18 on three Mediterranean islands (Sardinia, Crete, and Mallorca) through a blended Place-Based Education (PBE) model that integrates online learning with local, experiential activities. Forty-nine students completed a pre-assessment questionnaire measuring baseline marine ecosystem knowledge, sustainability-related behaviours, and attitudes toward the sea. Following three international exchanges involving the learning activities, roughly the same cohort of students completed post-activity surveys assessing self-perceived knowledge gains and intercultural interaction. Qualitative data from emotional mapping, field observations, and group reflections complemented the quantitative analysis. The results indicate substantial self-perceived increases in students’ understanding of marine ecosystems (+1.0 to +1.7 points on a 5-point scale), enhanced collaboration with international peers, and strengthened environmental awareness. Across all three sites, students applied their learning by co-designing proposals addressing local coastal challenges, demonstrating emerging civic responsibility and the ability to integrate scientific observations into real-world problem solving. These findings suggest that combining place-based education, citizen science, and participatory methods can effectively support the development of sustainability competences among youth in coastal contexts. This study contributes empirical evidence to the growing literature on education for sustainable development and highlights the value of blended, experiential, and intercultural approaches in promoting environmentally responsible behaviour. Full article

In modern e-learning environments, ensuring both student identity verification and concentration monitoring during online examinations has become increasingly important. This paper introduces a robust sequential framework that integrates Generative Adversarial Networks (GANs), Convolutional Neural Networks (CNNs) and fuzzy logic to achieve reliable face recognition and interpretable attentiveness assessment. Images from the Extended Yale B (cropped) dataset are preprocessed through grayscale normalization and resizing, while GANs generate synthetic variations in pose, illumination, and occlusion to enrich the training set and improve generalization. The CNN extracts discriminative facial features for identity recognition, and a fuzzy inference system transforms the CNN’s confidence scores into human-interpretable concentration levels. To stabilize learning and prevent overfitting, the model incorporates dropout regularization, batch normalization, and extensive data augmentation. Comprehensive evaluations using confusion matrices, ROC–AUC, and precision–recall analyses demonstrate an accuracy of 98.42%. The proposed framework offers a scalable and interpretable solution for secure and reliable online exam proctoring. Full article

Multi-Robot Task Allocation (MRTA) with spatiotemporal constraints presents significant challenges in environmental adaptability. Existing learning-based methods often overlook environmental spatial constraints, leading to spatial information distortion. To address this, we formulate the problem as an asynchronous Markov Decision Process over a directed heterogeneous graph and propose a novel heterogeneous graph neural network named the Edge-Enhanced Attention Network (E2AN). This network integrates a specialized encoder, the Edge-Enhanced Heterogeneous Graph Attention Network (E2HGAT), with an attention-based decoder. By incorporating edge attributes to effectively characterize path costs under spatial constraints, E2HGAT corrects spatial distortion. Furthermore, our approach supports flexible extension to diverse payload scenarios via node attribute adaptation. Extensive experiments conducted in simulated environments with obstructed maps demonstrate that the proposed method outperforms baseline algorithms in task success rate. Remarkably, the model maintains its advantages in generalization tests on unseen maps as well as in scalability tests across varying problem sizes. Ablation studies further validate the critical role of the proposed encoder in capturing spatiotemporal dependencies. Additionally, real-time performance analysis confirms the method’s feasibility for online deployment. Overall, this study offers an effective solution for MRTA problems with complex constraints. Full article

The Immersive Decentralized Digital space (IDDs), derived from blockchain technology and Massively Multiplayer Online Games (MMOGs), enables real-time multisensory interactions that support social connection under metaverse concepts. Although recognized as a technology with significant potential for educational innovation, IDDs remain underutilized in health professions education. Health profession students are often unaware of how IDDs’ features can be applied to their learning through in- or after-classroom activities. This study employs a quantitative research design to evaluate students’ perceptions of next-generation digital learning without any prior exposure to IDDs. An electronic survey was developed to examine four dimensions of learning facilitation: “Remote Learning” for capturing past experiences with digital competence during the COVID-19 era; “Digital Evolution,” reflecting preferences in utilizing digital spaces; “Interactive Communication” and “Knowledge Application” for applicability of IDDs in the health professions education. Statistical analyses revealed no significant differences in perceptions based on gender or major on all factors. Nevertheless, significant differences emerged based on nationality in “Digital Evolution”, “Interactive Communication”, and “Knowledge Application”, highlighting the influence of cultural and educational backgrounds on receptiveness to virtual learning environments. By recognizing the discrepancies and addressing barriers to digital inclusion, IDDs hold strong potential to enhance health professional learning experiences and educational outcomes. Full article

Adaptive filtering remains a cornerstone of modern signal processing but faces fundamental challenges when confronted with rapidly changing or nonlinear environments. This work investigates the integration of deep learning into adaptive-filter architectures to enhance tracking capability and robustness in non-stationary conditions. After reviewing and analyzing classical algorithms—LMS, NLMS, RLS, and a variable step-size LMS (VSS-LMS)—their theoretical stability and mean-square error behavior are formalized under a slow-variation system model. Comprehensive simulations using drifting autoregressive (AR(2)) processes, piecewise-stationary FIR systems, and time-varying sinusoidal signals confirm the classical trade-off between performance and complexity: RLS achieves the lowest steady-state error, at a quadratic cost, whereas LMS remains computationally efficient with slower adaptation. A stabilized VSS-LMS algorithm is proposed to balance these extremes; the results show that it maintains numerical stability under abrupt parameter jumps while attaining steady-state MSEs that are comparable to RLS (approximately 3 × 10⁻²) and superior robustness to noise. These findings are validated by theoretical tracking-error bounds that are derived for bounded parameter drift. Building on this foundation, a deep-learning-driven adaptive filter is introduced, where the update rule is parameterized by a neural function, $U_{\theta }$, that generalizes the classical gradient descent. This approach offers a pathway toward adaptive filters that are capable of self-tuning and context-aware learning, aligning with emerging trends in AI-augmented system architectures and next-generation computing. Future work will focus on online learning and FPGA/ASIC implementations for real-time deployment. Full article

Learner engagement is a pivotal factor affecting the effectiveness of Massive Open Online Courses (MOOCs), as it promotes collaborative learning environments. However, measuring the extent of learners’ contributions in MOOC learning forums presents challenges due to the complex nature of engagement and its variability. Given the limited research in this domain, further investigation is necessary. This study aims to address this gap by utilizing the Hidden Markov Model (HMM) to identify latent states of MOOC learners and improve their participation in learning forums. The study constructs a multidimensional observable signal sequence based on learner-generated post data from MOOC forums, with a particular focus on the widely attended course on a MOOC platform. To evaluate the predictive accuracy of HMM in forecasting learner contributions, the study employs several prominent prediction models for comparative analysis, including k-nearest neighbor, logistic regression, random forest, extreme gradient boosting tree, and the long short-term memory network. The results demonstrate that HMM provides superior accuracy in predicting learner contributions compared to other models. These findings not only validate the effectiveness of HMM but also offer significant insights and recommendations for enhancing forum management practices. This research represents a substantial advancement in addressing the challenges related to learner engagement in MOOC learning forums and underscores the potential benefits of employing the HMM approach in this context. Full article

The acceptance of, perceived advantages to, and skepticism toward the integration of artificial intelligence (AI) in undergraduate education is investigated in this paper. In this study, a total of 675 students from six Romanian universities answered a self-administered online questionnaire evaluating three main aspects: AI acceptance, AI benefits, and AI skepticism. While AI skepticism has a modest but substantial negative influence (β = −0.113, p = 0.001), results show that AI benefits favorably predict AI acceptance (β = 0.541, p = 0.001). Whereas AI skepticism negatively correlates with AI acceptance (r = −0.124, p = 0.001), correlational analysis reveals a high positive association between AI acceptance and AI benefits (r = 0.544, p = 0.001). Despite concerns about its limitations, the regression model suggests that students’ willingness to embrace AI in education is mostly driven by its perceived advantages. This explains 30.8% of the variance in AI acceptance (R² = 0.308, F(2, 641) = 142.909, p < 0.001). These results highlight the importance of techniques that improve perceived benefits while addressing uncertainty since they offer insightful analysis of student attitudes regarding artificial intelligence integration in higher education. By guiding policy decisions and educational activities meant to maximize AI-driven learning environments, this study adds to the current conversation on artificial intelligence adoption in education. Full article

Data prefetching is essential for modern file storage systems operating in large-scale cloud and data-intensive environments, where high performance increasingly depends on intelligent, adaptive mechanisms. Traditional rule-based methods and recently proposed machine learning-based techniques often struggle to cope with the complex and rapidly evolving data access patterns characteristic of big-data workloads. In this paper, we introduce an online, streaming machine learning (SML) approach for predictive data prefetching that retrieves useful data into the cache ahead of time. We present a novel online training framework that extracts features in real time and continuously updates streaming ML models to learn and adapt from large and dynamic access streams. Building on this framework, we design new SML-driven prefetching algorithms that decide when, how, and what data to prefetch into the cache with minimal overhead. Extensive experiments using production traces from Huawei Technologies Inc. and Google workloads from the SNIA IOTTA repository demonstrate that our intelligent policies consistently deliver the highest byte hits among competing approaches, achieving 97% prefetch byte precision and reducing data access latency by up to 2.8 times. These results show that streaming ML can deliver immediate performance gains and offers a scalable foundation for future adaptive storage systems. Full article