Search Results (75)

Educational chatbots are increasingly deployed to scaffold student learning, yet educators lack scalable ways to assess the cognitive depth of these dialogues in situ. Bloom’s taxonomy provides a principled lens for characterizing reasoning, but manual tagging of conversational turns is costly and difficult to scale for learning analytics. We present a reproducible high-confidence pseudo-labeling pipeline for multi-label Bloom classification of Socratic student–chatbot exchanges. The dataset comprises 6716 utterances collected from conversations between a Socratic chatbot and 34 undergraduate statistics students at Nanyang Technological University. From three chronologically selected workbooks with expert Bloom annotations, we trained and compared two labeling tracks: (i) a calibrated classical approach using SentenceTransformer (all-MiniLM-L6-v2) embeddings with one-vs-rest Logistic Regression, Linear SVM, XGBoost, and MLP, followed by per-class precision–recall threshold tuning; and (ii) a lightweight LLM track using GPT-4o-mini after supervised fine-tuning. Class-specific thresholds tuned on 5-fold cross-validation were then applied in a single pass to assign high-confidence pseudo-labels to the remaining unlabeled exchanges, avoiding feedback-loop confirmation bias. Fine-tuned GPT-4o-mini achieved the highest prevalence-weighted performance (micro-F1 $=0.814$), whereas calibrated classical models yielded stronger balance across Bloom levels (best macro-F1 $=0.630$ with Linear SVM; best classical micro-F1 $=0.759$ with Logistic Regression). Both model families reflect the corpus skew toward lower-order cognition, with LLMs excelling on common patterns and linear models better preserving rarer higher-order labels, while results should be interpreted as a proof-of-concept given limited gold labeling, the approach substantially reduces annotation burden and provides a practical pathway for Bloom-aware learning analytics and future real-time adaptive chatbot support. Full article

With the exponential growth of digital music, efficiently identifying key music relationship nodes in large-scale music knowledge graphs is crucial for enhancing music recommendation, emotion analysis, and genre classification. To address this challenge, we propose MUSIGAIN, a GATv2-based adaptive framework that combines graph robustness metrics with advanced graph neural network mechanisms for multi-relationship mining in heterogeneous music knowledge graphs. MUSIGAIN tackles three fundamental challenges: the prohibitive computational complexity of exact graph-robustness calculations, the limitations of traditional centrality measures in capturing semantic heterogeneity, and the over-smoothing problem in deep graph neural networks. The framework introduces three key innovations: (1) a layer-wise dynamic skipping mechanism that adaptively controls propagation depth based on third-order embedding stability, reducing computation by 30–40% while preventing over-smoothing; (2) the DiGRAF adaptive activation function that enables node-specific nonlinear transformations to capture semantic heterogeneity across different entity types; and (3) ranking-based optimization supervised by graph robustness metrics, focusing on relative importance ordering rather than absolute value prediction. Experimental results on four real-world music knowledge graphs (POP-MKG, ROCK-MKG, JAZZ-MKG, CLASSICAL-MKG) demonstrate that MUSIGAIN consistently outperforms existing methods in Top-5% node identification accuracy, achieving up to 96.78% while maintaining linear scalability to graphs with hundreds of thousands of nodes. MUSIGAIN provides an efficient, accurate, and interpretable solution for key node identification in complex heterogeneous graphs. Full article

With the emergence of large-scale vision-language pre-training (VLP) models, remote sensing (RS) image–text retrieval is shifting from global representation learning to fine-grained semantic alignment. This review systematically examines two mainstream representation paradigms—real-valued embedding and deep hashing—and analyzes how the evolution of RS datasets influences model capability, including multi-scale robustness, small object discriminability, and temporal semantic understanding. We further dissect three core challenges specific to RS scenarios: multi-scale semantic modeling, small object feature preservation, and multi-temporal reasoning. Representative architectures and technical solutions are reviewed in depth, followed by a critical discussion of their limitations in terms of generalization, evaluation consistency, and reproducibility. We also highlight the growing role of VLP-based models and the dependence of their performance on large-scale, high-quality image–text corpora. Finally, we outline future research directions, including RS-oriented VLP adaptation and unified multi-granularity evaluation frameworks. These insights aim to provide a coherent reference for advancing practical deployment and promoting cross-domain applications of RS image–text retrieval. Full article

The transition toward energy-efficient and smart buildings requires Digital Twins (DTs) that can couple real-time data with physics-based Building Energy Models (BEMs) for predictive and adaptive operation. Yet, despite rapid digitalisation, there remains a lack of practical guidance and real-world implementations demonstrating how calibrated BEMs can be effectively integrated into Building Management Systems (BMSs). This study addresses that gap by presenting a complete and reproducible end-to-end framework for embedding physics-based BEMs into operational DTs using two setups: (i) encapsulation as Functional Mock-up Units (FMUs) and (ii) containerisation via Docker. Both approaches were deployed and tested in a real educational building in Cáceres (Spain), equipped with a LoRaWAN-based sensing and actuation infrastructure. A systematic comparison highlights their respective trade-offs: FMUs offer faster execution but limited weather inputs and higher implementation effort, whereas Docker-based workflows provide full portability, scalability, and native interoperability with Internet of Things (IoT) and BMS architectures. To enable real-time operation, a surrogate modelling framework was embedded within the Docker architecture to replicate the optimisation logic of the calibrated BEM and generate predictive blind control schedules in milliseconds—bypassing simulation overhead and enabling continuous actuation. The combined Docker + surrogate setup achieved 10–15% heating energy savings during winter operation without any HVAC retrofit. Beyond the case study, this work provides a step-by-step, in-depth guideline for practitioners to integrate calibrated BEMs into real-time control loops using existing toolchains. The proposed approach demonstrates how hybrid physics- and data-driven DTs can transform building management into a scalable, energy-efficient, and operationally deployable reality. Full article

Accurate depth estimation is fundamental for underwater applications such as robotics and marine exploration. However, underwater imaging suffers from severe degradation due to light attenuation, scattering, and geometric distortion, which is compounded by the scarcity of real stereo data. To address these challenges, we propose Joint Dual-Branch Denoising (JDBD), which is a plug-in framework embedded within dual-branch depth estimation networks. JDBD performs task-aware denoising via bidirectional refinement between a monocular and a stereo pathway: the monocular branch combines Adaptive White Balance and a Red Inverse Channel Prior for color correction and haze suppression, while the stereo branch applies Joint Bilateral Filtering to reduce scattering and preserve edges. Trained on the synthetic UWStereo dataset and evaluated on the real-world SQUID dataset as well as a subset of UWStereo, JDBD achieves high depth estimation accuracy and visual fidelity in underwater scenes, demonstrating robust and adaptable performance across diverse conditions. Full article

Recent text to image systems can synthesize Islamic heritage elements with high visual fidelity, but their outputs rarely translate into fabricable geometry or integrate into interiors without substantial redrawing. We present an end-to-end workflow that links historically grounded precedent retrieval, controllable tileable generation, semantic segmentation and vectorization, and geometry-aware mapping into Computer-Aided Design (CAD) environments. Contributions include the following: (i) a license-audited dataset schema and a retrieval classifier for common Islamic motif families and architectural elements; (ii) precedent retrieval via a ResNet 50 and Vision Transformer (ViT) embedding pipeline; (iii) a Low-Rank Adaptation (LoRA) tuned diffusion model that generates tileable motifs with motif/region controls; (iv) a raster-to-vector pipeline that enforces curve closure and minimum feature widths for CNC/laser fabrication; and (v) a rubric and domain metrics (symmetry coherence, seam/tileability error, spline closure and junction valence, UV distortion, feature width compliance) that quantify “depth of integration” beyond surface texture. Quantitative metrics and blinded expert ratings compare the workflow against strong parametric baselines, while scripts translate images to fabrication-ready vectors/solids across walls, ceilings, partitions, floors, and furniture. Cultural safeguards cover calligraphy handling, regional balance audits, and provenance/credit. The workflow advances heritage-aware generative design by carrying imagery across the last mile into buildable detail and by providing practical checklists for adoption in interior architecture and conservation. Full article

The operation of modern power networks is increasingly exposed to overlapping climate extremes and volatile system conditions, making it essential to adopt scheduling approaches that are resilient as well as economical. In this study, a two-stage stochastic formulation is advanced, where indicators of system adaptability are embedded directly into the optimization process. The objective integrates standard operating expenses—generation, reserve allocation, imports, responsive demand, and fuel resources—with a Conditional Value-at-Risk component that reflects exposure to rare but damaging contingencies, such as extreme heat, severe cold, drought-related hydro scarcity, solar output suppression from wildfire smoke, and supply chain interruptions. Key adaptability dimensions, including storage cycling depth, activation speed of demand response, and resource ramping behavior, are modeled through nonlinear operational constraints. A stylized test system of 30 interconnected areas with a 46 GW demand peak is employed, with more than 2000 climate-informed scenarios compressed to 240 using distribution-preserving reduction techniques. The results indicate that incorporating risk-sensitive policies reduces expected unserved demand by more than 80% during compound disruptions, while the increase in cost remains within 12–15% of baseline planning. Pronounced spatiotemporal differences emerge: evening reserve margins fall below 6% without adaptability provisions, yet risk-adjusted scheduling sustains 10–12% margins. Transmission utilization curves further show that CVaR-based dispatch prevents extreme flows, though modest renewable curtailment arises in outer zones. Moreover, adaptability provisions promote shallower storage cycles, maintain an emergency reserve of 2–3 GWh, and accelerate the mobilization of demand-side response by over 25 min in high-stress cases. These findings confirm that combining stochastic uncertainty modeling with explicit adaptability metrics yields measurable gains in reliability, providing a structured direction for resilient system design under escalating multi-hazard risks. Full article

Near-lossless coding schemes traditionally rely on uniform quantization to control the maximum absolute error ($L_{\infty }$ norm) of residual signals, often assuming a parametric model for the source distribution. This paper introduces a novel design framework for non-uniform, entropy-aware $L_{\infty }$-oriented scalar quantizers that leverages a tight and differentiable approximation of the $L_{\infty }$ distortion metric and does not require any parametric density function formulations. The framework is evaluated on both synthetic parametric sources and real-world medical depth map video datasets. For smoothly decaying distributions, such as the continuous Laplacian or discrete two-sided geometric distributions, the proposed method naturally converges to near-uniform quantizers, consistent with theoretical expectations. In contrast, for sparse or irregular sources, the algorithm produces highly non-uniform bin allocations that adapt to the local distribution structure and improve rate-distortion efficiency. When embedded in a residual-based near-lossless compression scheme, the resulting codec consistently outperforms versions equipped with uniform or piecewise-uniform quantizers, as well as state-of-the-art near-lossless schemes such as JPEG-LS and CALIC. Full article

With the rapid development of autonomous vehicles and intelligent transportation systems, vehicle detection and distance estimation have become critical technologies for ensuring driving safety. However, real-world in-vehicle environments impose strict constraints on computational resources, making it impractical to deploy high-end GPUs. This implies that even highly accurate algorithms, if unable to run in real time on embedded platforms, cannot fully meet practical application demands. Although existing deep learning-based detection and stereo vision methods achieve state-of-the-art accuracy on public datasets, they often rely heavily on massive computational power and large-scale annotated data. Their high computational requirements and limited cross-scenario generalization capabilities restrict their feasibility in real-time vehicle-mounted applications. On the other hand, traditional algorithms such as Semi-Global Block Matching (SGBM) are advantageous in terms of computational efficiency and cross-scenario adaptability, but when used alone, their accuracy and robustness remain insufficient for safety-critical applications. Therefore, the motivation of this study is to develop a stereo vision-based collision warning system that achieves robustness, real-time performance, and computational efficiency. Our method is specifically designed for resource-constrained in-vehicle platforms, integrating a lightweight YOLOv8n detector with SGBM-based depth estimation. This approach enables real-time performance under limited resources, providing a more practical solution compared to conventional deep learning models and offering strong potential for real-world engineering applications. Full article

Thermally activated building systems (TABS) rely on high thermal mass materials, such as concrete, which perform well thermally but have a high carbon footprint. This study systematically investigates the thermal behavior of bio-based materials—spruce, pine, beech, and oak—in TABS using numerical simulations, comparing them with conventional and hybrid materials like concrete and clay. A total of 120 variants were simulated with different pipe diameters, spacing, embedment depths, and inlet temperatures. Thermal properties, particularly thermal conductivity and specific heat capacity, significantly influenced component activation efficiency. Concrete exhibited a characteristic cooling time of 71 h at an inlet temperature of 26 °C (pipe diameter 16 mm), while pine reached 80 h under the same conditions. The use of capillary tube mats extended the cooling times to 75 h for concrete and 92 h for pine. Although concrete provides the best thermal performance, certain bio-based materials achieve comparable results under optimized conditions. Hybrid systems with mineral components offer additional potential for improvement. These findings demonstrate that ecologically sustainable component activation using bio-based materials is feasible with only moderate efficiency losses compared to mineral-based systems, provided system parameters are appropriately adapted. Full article

Fiber-reinforced polymer (FRP) laminates are widely used in both aviation and structural engineering, yet their implementation reflects fundamentally different paradigms. Aviation represents a fatigue-critical, certification-driven domain, while structural engineering emphasizes long-term durability and environmental resilience. These sectors were selected as conceptual extremes to explore how contrasting design philosophies, degradation mechanisms, and inspection strategies shape the performance and reliability of laminated FRP composites. Their approaches offer complementary insights: aviation contributes high-fidelity modeling and embedded monitoring, while structural engineering provides scalable inspection strategies and exposure-based degradation logic. Both sectors employ classical laminate theory and finite element modeling, but diverge in modeling depth and regulatory integration. This review synthesizes these contrasts based on 168 literature references, including 141 published between 2020 and 2025, reflecting recent developments in composite design, modeling, and inspection. It contributes to materials engineering by proposing hybrid modeling and inspection frameworks that integrate progressive damage simulation with durability-based design logic. By bridging the modeling precision of aviation with the environmental realism of structural engineering, this review outlines a pathway toward unified, sustainable, and adaptive engineering practices for laminated FRP composites. Full article

Background: Swallowing is a complex biomechanical process, and its impairment (dysphagia) poses major health risks for older adults. Current diagnostic methods such as videofluoroscopic swallowing (VFSS) and fiberoptic endoscopic evaluation of swallowing (FEES) are effective but invasive, resource-intensive, and unsuitable for continuous monitoring. This study proposes a novel end-to-end RGB–D framework for automated swallowing event localization in continuous video streams. Methods: The framework enhances the AdaTAD backbone through three key innovations: (i) finding the optimal strategy to integrate depth information to capture subtle neck movements, (ii) examining the best adapter design for efficient temporal feature adaptation, and (iii) introducing a Kolmogorov–Arnold Network (KAN) decoder that leverages Chebyshev polynomials for non-linear temporal modeling. Evaluation on a proprietary swallowing dataset comprising 641 clips and 3153 annotated events demonstrated the effectiveness of the proposed framework. We analysed and compared the modification strategy across designs of adapters, decoders, input channel combinations, regression methods, and patch embedding techniques. Results: The optimized configuration (VideoMAE + GRNConvNeXtAdapter + KAN + RGD + boundary regression + sinusoidal embedding) achieved an average mAP of 83.25%, significantly surpassing the baseline I3D + RGB + MLP model (61.55%). Ablation studies further confirmed that each architectural component contributed incrementally to the overall improvement. Conclusions: These results establish the feasibility of accurate, non-invasive, and automated swallowing event localization using depth-augmented video. The proposed framework paves the way for practical dysphagia screening and long-term monitoring in clinical and home-care environments. Full article

(1) This study aims to enhance the precision of ergonomic fitting in traditional shoe size selection by integrating literature and measured biometric data. (2) A correlation table between biometric features and shoe models was established, which was then embedded into a knowledge graph (KG) for visual, accurate recommendations. The experiment employed pressure sensors and depth cameras to collect biometric data from the foot and leg, evaluating the consistency of the system’s recommendations and user satisfaction. (3) The results indicate that the biometric-driven shoe recommendation system significantly outperforms traditional size-based systems in terms of stability and satisfaction. (4) The KG framework has notably improved ergonomic adaptability in the early prototype stage, offering a viable technological approach for intelligent shoe selection and holding significant potential for further optimization. Full article

This research is an in-depth qualitative case study of a historic homestead in the town of Čekiškė, located in Lithuania, through the lens of sustainability aesthetics and cultural ecology. The research addresses a gap in the literature where aesthetic expressions of sustainability are predominantly examined in urban settings, while rural hybrid environments, intertwining urban and traditional features, remain underexplored. The homestead, with architectural and landscape features dating back to the early 20th century, was studied across four temporal stages: the interwar period (1922–1946), the early Soviet period (1946–1976), late Soviet to post-independence (1976–2021), and the period of a proposed vision for its sustainable development (2025 and beyond). The theoretical framework developed and applied in this research combines four complementary approaches: (1) the cultural ecology model by J. Steward, (2) environmental ethics trends (egocentrism, homocentrism, biocentrism, ecocentrism), (3) the principles of biophilic design, and (4) the ecological aesthetics framework by M. DeKay. Data collection included continuous qualitative in-depth on-site observations, analysis of the relevant literature sources, archival documents and photographs, and the recording of information in photographs and drawings. The findings reveal nuanced and evolving aesthetic expressions of sustainability tied to cultural practices, land use, ownership attitudes, and environmental perception. While earlier periods of development of the analyzed homestead reflected utilitarian and homocentric relations with the environment, later stages showed increased detachment from ecological sensitivity, resulting in the degradation of both material and intangible heritage; future perspectives of the homestead being transformed into a private museum, actualizing heritage through sustainability aesthetics, were also presented. The study highlights the role of tacit knowledge and lived experience in shaping hybrid sustainable aesthetics and provides insights for design and landscape planning in rural and small town heritage contexts. The research reveals that sustainability aesthetics in rural hybrid spaces is shaped by a confluence of environmental adaptation, socio-cultural transitions, and embedded values. It argues for a more context-sensitive and historically aware approach to sustainability discourse, particularly in heritage conservation and rural development. Full article

Leaves are the key organs in photosynthesis and nutrient production, and leaf counting is an important indicator of banana plant health and growth rate. However, in complex orchard environments, leaves often overlap, the background is cluttered, and illumination varies, making accurate segmentation and detection challenging. To address these issues, we propose a lightweight banana leaf detection and counting method deployable on embedded devices, which integrates a space–depth-collaborative reasoning strategy with multi-scale feature enhancement to achieve efficient and precise leaf identification and counting. For complex background interference and occlusion, we design a multi-scale attention guided feature enhancement mechanism that employs a Mixed Local Channel Attention (MLCA) module and a Self-Ensembling Attention Mechanism (SEAM) to strengthen local salient feature representation, suppress background noise, and improve discriminability under occlusion. To mitigate feature drift caused by environmental changes, we introduce a task-aware dynamic scale adaptive detection head (DyHead) combined with multi-rate depthwise separable dilated convolutions (DWR_Conv) to enhance multi-scale contextual awareness and adaptive feature recognition. Furthermore, to tackle instance differentiation and counting under occlusion and overlap, we develop a detection-guided space–depth position modeling method that, based on object detection, effectively models the distribution of occluded instances through space–depth feature description, outlier removal, and adaptive clustering analysis. Experimental results demonstrate that our YOLOv8n MDSD model outperforms the baseline by 2.08% in mAP50-95, and achieves a mean absolute error (MAE) of 0.67 and a root mean square error (RMSE) of 1.01 in leaf counting, exhibiting excellent accuracy and robustness for automated banana leaf statistics. Full article