Search Results (621)

Knowledge tracing (KT), a core educational data mining task, models students’ evolving knowledge states to predict future learning. In online education systems, the exercises are numerous, but they are typically associated with only a few concepts. However, existing models rarely integrate exercise information with high-order exercise–concept correlations, focusing solely on optimizing models’ final predictive performance. To address these limitations, we propose the Hypergraph-Driven High-Order Knowledge Tracing with a Dual-Gated Dynamic Mechanism (HGKT), a novel framework that (1) captures correlations between exercises and concepts through a two-layer hypergraph convolution; (2) integrates hypergraph-driven exercise embedding and temporal features (answer time and interval time) to characterize learning behavioral dynamics; and (3) designs a learning layer and a forgetting layer, with the dual-gating mechanism dynamically balancing their impacts on the knowledge state. Experiments on three public datasets demonstrate that the proposed HGKT model achieves superior predictive performance compared to all baselines. On the longest interaction sequence dataset, ASSISChall, HGKT improves prediction AUC by least 1.8%. On the biggest interaction records dataset, EdNet-KT1, it maintains a state-of-the-art AUC of 0.78372. Visualization analyses confirm its interpretability in tracing knowledge state evolution. These results validate HGKT’s effectiveness in modeling high-order exercise–concept correlations while ensuring practical adaptability in real-world online education platforms. Full article

This study explores the potential of craft-based activities in the context of special education, focusing on a papier mâché sculpting workshop implemented at the Special Kindergarten of Komotini, Greece, as part of the Horizon Europe Craeft project. The initiative aimed to assess how such creative activities could enhance the learning experience of children with intellectual and motor impairments, foster socialization, and develop fine motor skills. With reference to literature in art therapy, craft education, and inclusive pedagogy, the study applied a mixed-methods approach combining observation, visual analysis, and a survey. The findings indicate that, despite varied levels of participation based on individual needs, all students engaged meaningfully with the materials and activities. School professionals observed increased student engagement, emotional comfort, and communication, while also identifying the activity as well adapted and replicable in similar contexts. The results highlight the value of crafts in special education, not only as a sensory and cognitive stimulus but also as a means of fostering inclusion and self-expression. The study concludes with a call for further research into the role of tactile materials and hand gestures in relation to specific impairments. Full article

The present study focuses on investigating the contribution of puppetry as a pedagogical and psychosocial tool in special education, addressing the literature gap in the systematic documentation of the experiences of special education teachers, concerning its use in daily teaching practice. The main objective is to capture the way in which puppetry enhances the learning, social and therapeutic support of students with complex educational and psychosocial needs. The study employs a qualitative phenomenological approach, conducting semi-structured interviews with eleven special education teachers who integrate puppetry into their teaching. Qualitative data were analyzed using thematic analysis. The findings highlight that puppetry significantly enhances cognitive function, concentration, memory and language development, while promoting the active participation, cooperation, social inclusion and self-expression of students. In addition, the use of the puppet acts as a means of psycho-emotional empowerment, supporting positive behavior and helping students cope with stress and behavioral difficulties. Participants identified peer support, material adequacy and training as key factors for effective implementation, while conversely, a lack of resources and time is cited as a key obstacle. The integration of puppetry in everyday school life seems to ameliorate a more personalized, supportive and experiential learning environment, responding to the diverse and complex profiles of students attending special schools. Continuous training for teachers, along with strengthening the collaboration between the arts and special education, is essential for the effective use of puppetry in the classroom. Full article

Forest fire detection is an essential application in environmental surveillance since wildfires cause devastating damage to ecosystems, human life, and property every year. The effective and accurate detection of fire is necessary to allow for timely response and efficient management of disasters. Traditional techniques for fire detection often experience false alarms and delayed responses in various environmental situations. Therefore, developing robust, intelligent, and real-time detection systems has emerged as a central challenge in remote sensing and computer vision research communities. Despite recent achievements in deep learning, current forest fire detection models still face issues with generalizability, lightweight deployment, and accuracy trade-offs. In order to overcome these limitations, we introduce a novel technique (FireNet-KD) that makes use of knowledge distillation, a method that maps the learning of hard models (teachers) to a light and efficient model (student). We specifically utilize two opposing teacher networks: a Vision Transformer (ViT), which is popular for its global attention and contextual learning ability, and a Convolutional Neural Network (CNN), which is esteemed for its spatial locality and inductive biases. These teacher models instruct the learning of a Swin Transformer-based student model that provides hierarchical feature extraction and computational efficiency through shifted window self-attention, and is thus particularly well suited for scalable forest fire detection. By combining the strengths of ViT and CNN with distillation into the Swin Transformer, the FireNet-KD model outperforms state-of-the-art methods with significant improvements. Experimental results show that the FireNet-KD model obtains a precision of 95.16%, recall of 99.61%, F1-score of 97.34%, and mAP@50 of 97.31%, outperforming the existing models. These results prove the effectiveness of FireNet-KD in improving both detection accuracy and model efficiency for forest fire detection. Full article

Semi-supervised object detection (SSOD) in remote sensing is challenged by the accumulation of pseudo-label errors in complex scenes with dense objects and high intra-class variability. While teacher–student frameworks enable learning from unlabeled data, erroneous pseudo-labels such as false positives and missed detections can be reinforced over time, which degrades model performance. To address this issue, we propose the Error-Mitigation Teacher (EMT), a unified framework designed to suppress error propagation during SSOD training. EMT consists of three lightweight modules. First, the Adaptive Pseudo-Label Filtering (APLF) module removes noisy pseudo boxes via a second-stage RCNN and adjusts class-specific thresholds through dynamic confidence filtering. Second, the Confidence-Based Loss Reweighting (CBLR) module reweights training loss by evaluating the teacher model’s ability to reconstruct its own pseudo-labels, using the resulting loss as an indicator of label reliability. Third, the Enhanced Supervised Learning (ESL) module improves class-level balance by adjusting supervised loss weights according to pseudo-label statistics. EMT demonstrates consistent performance gains over representative state-of-the-art SSOD methods on DOTA, DIOR, and SSDD datasets. Notably, EMT achieves a 2.9% absolute mAP₅₀ improvement on DIOR using only 10% of labeled data, without incurring additional inference cost. These results highlight EMT’s effectiveness in improving SSOD for remote sensing. Full article

The transformer-based deep learning approach represents the current state-of-the-art in machine translation (MT) research. Large-scale pretrained transformer models produce state-of-the-art performance across a wide range of MT tasks for many languages. However, such deep neural network (NN) models are often data-, compute-, space-, power-, and energy-hungry, typically requiring powerful GPUs or large-scale clusters to train and deploy. As a result, they are often regarded as “non-green” and “unsustainable” technologies. Distilling knowledge from large deep NN models (teachers) to smaller NN models (students) is a widely adopted sustainable development approach in MT as well as in broader areas of natural language processing (NLP), including speech, and image processing. However, distilling large pretrained models presents several challenges. First, increased training time and cost that scales with the volume of data used for training a student model. This could pose a challenge for translation service providers (TSPs), as they may have limited budgets for training. Moreover, CO₂ emissions generated during model training are typically proportional to the amount of data used, contributing to environmental harm. Second, when querying teacher models, including encoder–decoder models such as NLLB, the translations they produce for low-resource languages may be noisy or of low quality. This can undermine sequence-level knowledge distillation (SKD), as student models may inherit and reinforce errors from inaccurate labels. In this study, the teacher model’s confidence estimation is employed to filter those instances from the distilled training data for which the teacher exhibits low confidence. We tested our methods on a low-resource Urdu-to-English translation task operating within a constrained training budget in an industrial translation setting. Our findings show that confidence estimation-based filtering can significantly reduce the cost and CO₂ emissions associated with training a student model without drop in translation quality, making it a practical and environmentally sustainable solution for the TSPs. Full article

Advanced three-dimensional extended reality (XR) technologies are highly suitable for cultural heritage research and education. XR tools enable the creation of realistic virtual or augmented reality applications for curating and disseminating information about cultural artifacts and sites. Developing XR applications for cultural heritage requires interdisciplinary collaboration involving strong teamwork and soft skills to manage user requirements, system specifications, and design cycles. Given the diverse end-users, achieving high precision, accuracy, and efficiency in information management and user experience is crucial. Human–computer interaction (HCI) design and evaluation methods are essential for ensuring usability and return on investment. This article presents ten case studies of cultural heritage software projects, illustrating the interdisciplinary work between computer science and HCI design. Students from institutions such as the State University of New York (USA), Glasgow School of Art (UK), University of Granada (Spain), University of Málaga (Spain), Duy Tan University (Vietnam), Imperial College London (UK), Research University Institute of Communication & Computer Systems (Greece), Technical University of Košice (Slovakia), and Indiana University (USA) contributed to creating, assessing, and improving the usability of these diverse cultural heritage applications. The results include a structured typology of CH XR application scenarios, detailed insights into design and evaluation practices across ten international use cases, and a development framework that supports interdisciplinary collaboration and stakeholder integration in phygital cultural heritage projects. Full article

Retrieval practice, or the active recall of information from memory, is a highly effective learning strategy that strengthens memory and comprehension. This effect is robust and strongly backed by research in cognitive psychology. The health professions—including medicine, nursing, and dentistry—have widely embraced retrieval practice as a learning and study tool, particularly for course exams and high-stakes licensing exams. This state-of-the-art review examines the historical development, current applications, and future directions for the use of retrieval practice in health professions education. While retrieval-based learning has long been used informally in these fields, its formal recognition as a scientifically supported study method gained momentum in the early 2000s and then saw a surge in both research interest and curricular adoption between 2010 and 2025. This historical review explores the key factors driving this growth, such as its alignment with assessment-driven education and the increasing availability of third-party study resources that rely on retrieval practice as a guiding principle. Despite its proven benefits for learning, however, barriers persist to its adoption by students, including in the health professions. This article discusses strategies for overcoming these challenges and for enhancing retrieval practice integration into health professions curricula. Full article

This study investigates how lighting conditions influence emotional and neural responses in a standardized, simulated museum environment. A multimodal evaluation framework combining subjective and objective measures was used. Thirty-two international students assessed their viewing experiences using 14 semantic differential descriptors, while real-time EEG signals were recorded via the EMOTIV EPOC X device. Spectral energy analyses of the α, β, and θ frequency bands were conducted, and a θα energy ratio combined with γ coefficients was used to model attention and comfort levels. The results indicated that high illuminance (300 lx) and high correlated color temperature (4000 K) significantly enhanced both attention and comfort. Art majors showed higher attention levels than engineering majors during short-term viewing. Among four regression models, the backpropagation (BP) neural network achieved the highest predictive accuracy (R² = 88.65%). These findings provide empirical support for designing culturally inclusive museum lighting and offer neuroscience-informed strategies for promoting the global dissemination of traditional Chinese culture, further supported by retrospective interview insights. Full article

Background: The perceived importance of physical practice and its contribution to students’ academic success have evolved considerably throughout the history of the modern educational system. Aim: The purpose of this study was to understand the relationship between physical fitness (measured as VO₂max) and cognitive abilities (attention and concentration) and academic performance in different subjects: sciences, letters, language, arts, and physical education. Method: Fifty Spanish male students who participated in extracurricular sports activities (mean age (SD): 11.59 ± 1.30; range: 9–15 years) were included in the analysis. The 6 min walk test was used to assess physical fitness (6MWT), while for selective attention and concentration, the students completed the D2 test, which is usually considered to analyse the visual ability to select the most relevant stimulus of an exercise and ignore precisely the most irrelevant stimuli. Results: Correlation the individual contribution analyses revealed no significant associations between VO₂max and academic performance in sciences (r = 0.04, p = 0.77), humanities (r = 0.00, p = 0.98), language (r = 0.03, p = 0.83), or arts (r = 0.04, p = 0.76). Similarly, no relationship was found between VO₂max and overall academic performance (r = 0.10, p = 0.46), or cognitive abilities. However, a small positive correlation was observed between VO₂max and physical education scores. Conclusions: Physical fitness showed no significant association with cognitive abilities or academic performance in most subjects, although a small positive correlation with physical education scores was observed. These findings emphasise the importance of promoting physical activity for its health and physical benefits. However, future research should explore broader cognitive outcomes and include more diverse and representative samples. Full article

This paper presents the implementation of a modern approach for automatic measurement, data acquisition, and processing using custom-developed software based on the ARDUINO version: 2.3.6 platform. State-of-the-art sensing elements are employed for enhanced precision and reliability. The obtained results are graphically visualized for comprehensive analysis. The ARDUINO microcontroller and its associated open-source programming environment are primarily designed for general-purpose users rather than specialized industrial applications. The study focuses on the experimental investigation of the characteristics of a Peltier element. The interdependence between current, voltage, internal resistance, and temperature differential is examined in detail. The findings concerning efficiency analysis are intended to support students across various engineering disciplines during their educational process. Full article

Multimodal emotion recognition has emerged as a prominent field in affective computing, offering superior performance compared to single-modality methods. Among various physiological signals, EEG signals and EOG data are highly valued for their complementary strengths in emotion recognition. However, the practical application of EEG-based approaches is often hindered by high costs and operational complexity, making EOG a more feasible alternative in real-world scenarios. To address this limitation, this study introduces a novel framework for multimodal knowledge distillation, designed to improve the practicality of emotion decoding while maintaining high accuracy, with the framework including a multimodal fusion module to extract and integrate interactive and heterogeneous features, and a unimodal student model structurally aligned with the multimodal teacher model for better knowledge alignment. The framework combines EEG and EOG signals into a unified model and distills the fused multimodal features into a simplified EOG-only model. To facilitate efficient knowledge transfer, the approach incorporates a dynamic feedback mechanism that adjusts the guidance provided by the multimodal model to the unimodal model during the distillation process based on performance metrics. The proposed method was comprehensively evaluated on two datasets based on EEG and EOG signals. The accuracy of the valence and arousal of the proposed model in the DEAP dataset are 70.38% and 60.41%, respectively. The accuracy of valence and arousal in the BJTU-Emotion dataset are 61.31% and 60.31%, respectively. The proposed method achieves state-of-the-art classification performance compared to the baseline method, with statistically significant improvements confirmed by paired t-tests (p < 0.05), and the framework effectively transfers knowledge from multimodal models to unimodal EOG models, enhancing the practicality of emotion recognition while maintaining high accuracy, thus expanding the applicability of emotion recognition in real-world scenarios. Full article

Knowledge tracing (KT) models learners’ evolving knowledge states to predict future performance, serving as a fundamental component in personalized education systems. However, existing methods suffer from data sparsity challenges, resulting in inadequate representation quality for low-frequency knowledge concepts and inconsistent modeling of students’ actual knowledge states. To address this challenge, we propose Dual-Encoder Contrastive Knowledge Tracing (DECKT), a contrastive learning framework that improves knowledge state representation under sparse data conditions. DECKT employs a momentum-updated dual-encoder architecture where the primary encoder processes current input data while the momentum encoder maintains stable historical representations through exponential moving average updates. These encoders naturally form contrastive pairs through temporal evolution, effectively enhancing representation capabilities for low-frequency knowledge concepts without requiring destructive data augmentation operations that may compromise knowledge structure integrity. To preserve semantic consistency in learned representations, DECKT incorporates a graph structure constraint loss that leverages concept–question relationships to maintain appropriate similarities between related concepts in the embedding space. Furthermore, an adversarial training mechanism applies perturbations to embedding vectors, enhancing model robustness and generalization. Extensive experiments on benchmark datasets demonstrate that DECKT significantly outperforms existing state-of-the-art methods, validating the effectiveness of the proposed approach in alleviating representation challenges in sparse educational data. Full article

Knowledge distillation (KD) is crucial for remote sensing image (RSI) classification, particularly as the operating environment in remote sensing is often constrained by hardware limitations. However, prior research has not fully addressed the challenge of leveraging KD to develop lightweight, high-accuracy models for RSI classification. A key issue is the sparse distribution of training data, which often results in asymmetry within the data. This asymmetry impedes the transfer of prior knowledge during the distillation process, diminishing the overall efficacy of KD techniques. To overcome this challenge, we propose a novel, symmetry-enhanced approach that augments the logit-based KD process, improving its effectiveness and efficiency for RSI classification. Our method is distinguished by three core innovations: a symmetrically generative algorithm to enhance the symmetry of the training data, an efficient algorithm for constructing a robust teacher ensemble model, and a quantitative technique for feature alignment. Rigorous evaluations on three benchmark datasets demonstrate that our method outperforms 14 existing KD-based approaches and 30 other state-of-the-art methods. Specifically, the student model trained with our approach achieves accuracy improvements of up to 22.5% while reducing the model size and inference time by as much as 96% and 88%, respectively. In conclusion, this research makes a significant contribution to RSI classification by introducing an efficient and effective data symmetry-driven method to enhance the knowledge transferring efficiency of the logit-based KD process. Full article

The Draw-A-Person Intellectual Ability test for children, adolescents, and adults is a widely used tool in psychology for assessing intellectual ability. This test relies on human drawings for initial raw scoring, with the subsequent conversion of data into IQ ranges through manual procedures. However, this manual scoring and IQ assessment process can be time-consuming, particularly for busy psychologists dealing with a high caseload of children and adolescents. Presently, DAP-IQ screening continues to be a manual endeavor conducted by psychologists. The primary objective of our research is to streamline the IQ screening process for psychologists by leveraging deep learning algorithms. In this study, we utilized the DAP-IQ manual to derive IQ measurements and categorized the entire dataset into seven distinct classes: Very Superior, Superior, High Average, Average, Below Average, Significantly Impaired, and Mildly Impaired. The dataset for IQ screening was sourced from primary to high school students aged from 8 to 17, comprising over 1100 sketches, which were subsequently manually classified under the DAP-IQ manual. Subsequently, the manual classified dataset was converted into digital images. To develop the artificial intelligence-based models, various deep learning algorithms were employed, including Convolutional Neural Network (CNN) and state-of-the-art CNN (Transfer Learning) models such as Mobile-Net, Xception, InceptionResNetV2, and InceptionV3. The Mobile-Net model demonstrated remarkable performance, achieving a classification accuracy of 98.68%, surpassing the capabilities of existing methodologies. This research represents a significant step towards expediting and enhancing the IQ screening for psychologists working with diverse age groups. Full article