Purpose This study aims to develop a hybrid model integrating the decision-making trial and evaluation laboratory (DEMATEL) method and System Dynamics (SD) to analyze how green variables (green training, environmental awareness and green practice adoption) enhance human resource capabilities (HRCs). By identifying causal relationships among ten key human resource (HR) factors and simulating their dynamic interactions, the research provides insights into strengthening organizational sustainability. The model offers a strategic framework for policymakers and managers to align HR development with environmental objectives and build adaptive, resilient workforces. Design/methodology/approach This study used an integrated framework combining the DEMATEL and SD. Expert assessments from industrial specialists were collected to identify and analyze causal relationships among key HR capability factors. These relationships were then incorporated into a SD model to simulate long-term behaviors under multiple green scenarios, enabling both static and dynamic evaluation of human resource capabilities in sustainable industrial contexts. Findings The results reveal that green training, environmental awareness and the adoption of green practices significantly enhance critical Human Resource Capabilities, particularly adaptability, professionalism, safety awareness and responsibility. These factors not only improve individual competencies but also generate reinforcing feedback loops within the organizational system. The dynamic simulations further show that combined green interventions accelerate capability growth, promoting long-term organizational resilience, innovation and alignment with sustainability objectives in industrial environments. Research limitations/implications This research primarily relies on simulation data derived from expert assessments, which limits its external validity. Although sensitivity tests and scenario comparisons were performed, the absence of empirical organizational data constrains the generalizability of findings. Future studies should incorporate real-world data sets across diverse industries to validate the model, refine causal assumptions and explore additional HR and green variables for broader application in sustainable workforce management. Practical implications The study provides actionable insights for managers and policymakers by showing how green training, environmental awareness and eco-friendly practices can be systematically embedded into HR strategies. By adopting this integrated approach, organizations can enhance workforce adaptability, responsibility and professionalism while aligning HR policies with sustainability objectives. This framework supports long-term competitiveness, improved employee engagement and the development of resilient workforces capable of responding to dynamic environmental and industrial challenges. Social implications This study highlights how embedding green initiatives within HR practices fosters eco-friendly behaviors, ethical responsibility and social awareness among employees. By strengthening teamwork, professionalism and safety consciousness, organizations can create a culture that values environmental stewardship and collective resilience. Such practices not only contribute to reduced ecological impact but also enhance social sustainability, ensuring that industrial growth is aligned with broader societal well-being and long-term environmental responsibility. Originality/value The study offers a novel and replicable framework that systematically integrates human resource management practices with organizational sustainability goals, emphasizing feedback loops, continuous learning and adaptive strategies to enhance both employee performance and long-term environmental and social impact.

This qualitative library research examines the ethical foundations and learning strategies for seekers of knowledge as articulated by Muhammad bin Ṣāliḥ Al-Utsaimin in Kitab al-Ilm. Using content analysis, the study systematically codes thematic units related to intention, discipline, reverence for scholars, epistemic verification, and pedagogical method. The findings reveal twelve interconnected ethical principles that function not merely as moral guidance but as an integrated epistemic framework. These include sincerity of intention, commitment to acting upon knowledge, respect for scholarly authority, patience in learning, intellectual humility, and verification of information. Complementing these ethical foundations are practical strategies such as prioritizing foundational knowledge, structured study routines, memorization and revision (murāja’ah), guided mentorship (mulāzamah), and selective academic companionship. The study argues that Al-Utsaimin’s framework represents a holistic model of Islamic pedagogy in which ethics and methodology are inseparable. In contemporary digital learning environments characterized by information overload and academic shortcuts, this framework offers a morally grounded and methodologically rigorous alternative capable of strengthening academic integrity and character formation among Generation Z learners.

This study aims to analyze and describe diagnostic learning strategies in the internalization process of a love-based curriculum at Madrasah Aliyah Al-Ittifaqiah. A love-based curriculum is understood as an educational approach that emphasizes compassion, empathy, responsibility, and spirituality as the foundation for students’ character development. The primary challenge in its implementation lies in ensuring that values are not confined to cognitive understanding but are manifested in students’ attitudes and behaviors.This research employed a qualitative approach with a case study design. Data were collected through participatory observation, in-depth interviews with the principal and teachers, and analysis of instructional documents. Data analysis was conducted through systematic data reduction, display, and conclusion drawing. The findings reveal that diagnostic learning strategies were implemented through initial mapping of students’ characteristics, identification of learning needs and socio-religious backgrounds, and continuous formative assessment. Teachers utilized diagnostic results to design contextual, reflective, and exemplary-based learning processes.The study concludes that diagnostic learning plays a significant role in strengthening the internalization of love-based values through habituation, educational dialogue, and integration of values into classroom activities. It recommends enhancing teachers’ competencies in diagnostic assessment and developing structured character evaluation instruments to support the sustainability of love-based curriculum implementation in madrasah education.

Hyperspectral imaging (HSI) captures measurements at multiple wavelengths across the electromagnetic spectrum, providing information that improves material segmentation and classification beyond RGB imagery. While HSI devices often acquire a large number of spectral bands, this increases both cost and acquisition time. However, not all bands contribute equally to task-specific performance. We propose a deep spectral band selection (DSBS) framework for HSI tasks. Unlike methods that preserve non-task-specific information, DSBS identifies the most informative bands for a given task by jointly training a fully differentiable band selector and a neural network in an end-to-end (E2E) learning scheme. The selection is guided by a bin function and an ℓ p -norm regularizer to reach a target number of bands. Experiments on segmentation and classification show that DSBS outperforms state-of-the-art machine learning and deep learning methods. Results show that DSBS outperforms state-of-the-art E2E methods for HSI classification by about 8% across the evaluated metrics and yields an improvement of 1% over using all spectral bands for material segmentation. Additionally, we validate DSBS in a testbed implementation, starting from full-spectrum images (301 bands). The end-to-end training converges to 10 wavelengths; under cross-validation, considering only these 10 bands yields competitive cocoa-bean classification, with an overall accuracy of 76.40%, retaining approximately 95.5% of the accuracy observed in the 10-band end-to-end evaluation (80.01%) while reducing acquisition speed by a factor of 30 times.

Deaf and hard of hearing (DHH) children may develop compensatory strategies in language learning due to auditory limitations, such as using speechreading to process spoken information. However, it remains unclear whether similar visual compensatory mechanisms are employed by DHH children when acquiring written languages, especially in visually complex logographic systems such as Chinese. This study examined the characteristics of speechreading as well as phonological and visual-orthographic processing skills in Chinese reading among DHH children, utilizing both age- and reading level-matched designs. A total of 24 DHH children who used sign language participated in the present study, alongside 24 age-matched and 24 reading level-matched hearing peers. All participants completed a battery of assessments on reading and related cognitive skills, including Chinese character reading, sentence comprehension, text comprehension, phonological awareness, rapid naming, speechreading comprehension, and orthographic and visual processing. Whereas DHH children performed significantly worse than age- and reading level-matched hearing children on phonological awareness and rapid naming, they performed significantly better than the two hearing groups on speechreading comprehension and visual search tasks. In addition, DHH children also performed comparably to the two hearing groups on orthographic skills. Correlation analyses revealed that, for DHH children, higher phonological awareness was significantly correlated with better Chinese character reading and that faster rapid naming and better speechreading at the single-word level were significantly correlated with increased text comprehension. Additionally, the hit rate of real characters and visual processing showed a trend toward correlation with Chinese character reading. These findings suggest that, despite challenges in auditory and phonological processing, DHH children may resort to visual compensatory strategies to facilitate their reading development. https://doi.org/10.23641/asha.31337374.

This study investigates the development of productive learning habits through temporal regularity in learning activities. Building effective habits involves self-regulated learning (SRL) strategies, particularly in time management, which are critical for learners to regulate their behaviors and optimize their productivity. While Learning Analytics (LA) techniques have been employed to monitor habitual behaviors and provide long-term support, few of them attended to learners’ decisions on which habit to build when they try to find their optimal time for learning. To address this gap, we designed an algorithm that generates personalized recommendations for optimal learning time slots based on learning log data. Our findings reveal that these recommendations can increase learners’ awareness of productive time slots, guide them in aligning their behaviors with their goals, and support the development of sustainable learning habits. The study also highlights the implications for K-12 learners who often lack specific time management skills, and educators, who can leverage such tools to provide structured guidance and targeted feedback. By integrating adaptive learning systems and personalized recommendations, this study contributes to advancing SRL support within technology–enhanced learning environments, offering practical insights for improving time management, goal setting, and overall learning productivity.

The software and data in this repository are a snapshot of the software and data that were used in the research reported on in the paper When Multimodal Interactions Impair Prediction: A Novel Regularized Deep Learning Strategy by Gang Chen, Shuaiyong Xiao, Chenghong Zhang, and Huimin Zhao.

Bladder cancer is one of the most prevalent malignancies of the urinary system and is associated with high morbidity and mortality. With advances in medical image analysis, deep learning has shown promise for automated bladder cancer classification using magnetic resonance imaging (MRI). However, clinical deployment remains challenging due to substantial inter-center distributional discrepancies and limited feature discriminability between non-muscle-invasive bladder cancer (NMIBC) and muscle-invasive bladder cancer (MIBC). To address these challenges, we propose a Domain-Adaptive Deep Contrastive Network (DADCNet) for MRI-based bladder cancer classification. The proposed framework jointly incorporates source- and target-domain samples during feature learning to obtain domain-invariant yet discriminative representations, thereby improving cross-center generalization. In addition, a deep contrastive learning strategy is introduced to enhance inter-class separability and intra-class compactness, leading to more robust classification. Experiments conducted on a multi-center bladder cancer MRI dataset demonstrate that DADCNet consistently outperforms existing convolutional neural network- and Transformer-based methods, achieving an accuracy of 0.955, an F1-score of 0.955, and an area under the curve of 0.991.

Given the substantial body of research indicating that feedback is essential for teaching and learning across disciplines, I propose a minimal empirically based notion of feedback that helps improve educational practices in philosophy. While current pedagogical approaches in philosophy often emerge from limited sources like personal experiences and peer discussions, it is important to acknowledge that philosophers have explored a variety of successful teaching and learning strategies based on empirical research. However, fundamental elements underlying all teaching and learning practices – such as feedback – remain insufficiently examined. Robust empirical evidence highlights feedback as one of the most critical factors influencing learning success across disciplines, including philosophy. By better understanding and incorporating feedback into philosophical pedagogy, we can significantly enhance the learning of philosophy students and the teaching practices of philosophy teachers.

Although active learning methods have been widely examined in educational research, the effect of the brainstorming method on learning outcomes in Islamic Religious Education (PAI) at the junior high school level, particularly in schools located in non-urban areas, remains relatively underexplored. This study aimed to analyze the effect of the brainstorming method on students’ learning outcomes in the Islamic Religious Education subject at SMP N 1 Gunuang Omeh Subdistrict. A quantitative approach was employed with a quasi-experimental design in the form of a Static Group Comparison Design, involving 40 eighth-grade students selected through total sampling. Data were collected using an objective multiple-choice test and analyzed with the nonparametric Mann–Whitney U test after normality and homogeneity tests were conducted. The results showed a significant difference between the experimental and control groups (p = 0.003 < 0.05), where the experimental group taught using the brainstorming method obtained a mean score of 83.40 with an 85% mastery level, while the control group taught using conventional methods obtained a mean score of 74.10 with a 55% mastery level. These findings reinforce constructivist and social learning theories in the context of Islamic Religious Education and confirm that the implementation of active learning strategies based on brainstorming is effective in improving students’ learning outcomes. The study recommends that Islamic Religious Education teachers further integrate the brainstorming method into classroom practice as part of developing student-centered learning, and encourages subsequent research to examine the impact of this method on affective and psychomotor domains.

Artificial-intelligence-driven wayside monitoring has become a promising solution for early identification of railway wheel flats, enabling safer operations and more efficient maintenance planning. This study introduces a comparative investigation of supervised and unsupervised machine learning strategies for wheel flat identification, with particular emphasis on real-time applicability and sensor cost reduction. Support Vector Machines (SVMs) and k-means clustering are evaluated as representative supervised and unsupervised approaches using vibration data obtained from numerically simulated train–track interactions under realistic operating conditions, including train speeds of 120 km/h and 200 km/h and multiple wheel flat severities. A key contribution of this work is the proposal of a simplified supervised classification framework that directly exploits Auto-Regressive features extracted from rail-mounted accelerometers, eliminating the need for feature normalization and multi-sensor data fusion. This simplification significantly reduces computational effort, making the approach suitable for real-time deployment in operational railway environments. In parallel, a systematic sensitivity analysis is conducted to assess the influence of sensor placement and to identify the minimum sensor configuration required to achieve reliable damage classification. The outputs from the current study show that an SVM emerges with more accurate defect classification than the k-means clustering, allowing a wayside system with fewer sensors.

In the era of digital transformation and VUCA (volatility, uncertainty, complexity, ambiguity) environments, continuous organizational learning has become a strategic necessity to remain viable. While traditional classroom training was the standard, e-learning is increasingly adopted for its efficiency and flexibility in human resource development. However, doubts remain regarding the comparative effectiveness of digital versus face-to-face formats, specifically concerning dropout rates and social isolation. This study investigates the specific advantages of each format and the critical role of organizational learning culture in ensuring success. Based on a systematic literature review, this paper analyses theoretical foundations and practical applications to assess both learning modalities. It conducts a comparative assessment to determine their suitability for different organizational contexts and skill requirements. The analysis reveals that while e-learning excels in flexibility and factual knowledge transfer, face-to-face training is superior for social interaction and developing practical skills. Effectiveness depends less on the format itself and more on the didactic design and alignment with the organizational learning culture. These findings imply that organizations should move beyond an either/or approach towards blended learning strategies. To remain viable in the long term, they must promote Deutero learning (the ability to learn how to learn) within their cultural framework.

Background Tuberculosis (TB) remains a major global health threat, causing approximately 1.5 million deaths each year. Despite progress in treatment, 15%–20% of patients still experience treatment failure or relapse, highlighting the urgent need for precise predictive tools for early identification of high-risk patients. Current methods based on clinical parameters have limitations in prediction accuracy and revealing potential biological mechanisms. Methods This study developed and validated an innovative multi-omics integration prediction model. We retrospectively collected clinical data from 467 tuberculosis patients and integrated transcriptomic data from three independent public cohorts (GSE19491, GSE31312, GSE83456), involving 3,240 differentially expressed genes. Through advanced feature engineering and bioinformatics analysis, key features were selected. We systematically evaluated 12 machine learning algorithms and adopted an ensemble learning strategy to construct the final model. Model performance was evaluated through strict cross-validation and prospective validation cohorts. Results Clinical data analysis identified age, body mass index (BMI), and C-reactive protein (CRP) levels as significant predictors of treatment response. Transcriptomic analysis revealed 1,247 differentially expressed genes between responders and non-responders, enriched in immune response and metabolic pathways. Among the tested algorithms, the ensemble model based on Extra Trees performed the best, with an area under the curve (AUC) of 0.986, significantly superior to models using only clinical data (AUC = 0.850) or only genomic data (AUC = 0.820). Feature importance analysis confirmed CRP, specific gene features (such as DNA repair and interferon response pathways), age, and BMI as the most important predictors. External validation confirmed the model’s robustness (AUC = 0.972). Conclusion This study successfully developed a high-precision prediction model integrating clinical and genomics data, capable of early identification of high-risk patients with poor treatment response. The model demonstrates excellent prediction performance and generalization ability, providing a powerful tool for moving towards tuberculosis precision medicine, guiding individualized treatment strategies to improve patient prognosis and control the spread of drug resistance. Clinical Trial Registration https://www.chictr.org.cn/ , ChiCTR2300074328, 03/08/2023.

Massive open online courses (MOOCs) are well-known for offering flexible learning; however, they also pose challenges for English-as-a-foreign-language (EFL) learners, particularly in maintaining motivation and engagement in such self-paced learning online courses. This study explored the factors influencing EFL students' participation in a MOOC through the lens of Expectancy-Value Theory and Socio-Cognitive Theory. Employing a qualitative design, involving 31 EFL students from a private university in the Mekong Delta enrolled in a public speaking MOOC, the study collected data of 20 reflective journals and 12 semi-structured interviews from these participants. The data was analyzed using reflexive thematic analysis. The findings revealed that student engagement increased when learners expected success, perceived the course as valuable, and employed self-regulated learning strategies. Motivation was sustained through goal setting, time management, and persistence; however, language barriers, technological issues, and social isolation impeded learning. Despite these obstacles, students use goal setting, time management, and persistence to sustain motivation. Expectancy–value theory (EVT) explains how perceived value and expectations of success shape engagement, whereas social cognitive theory (SCT) emphasizes self-efficacy and self-regulation as key drivers of motivation. These findings can inform educators, curriculum designers, and policymakers in developing more supportive online learning environments. Alongside recommending inclusive instructional strategies, including linguistic support, interactive content, and community-building, the study urges research into the long-term effects of SRL strategies and the effectiveness of SRL-based training for EFL learners in varied MOOC contexts.

Undergraduate EFL students across diverse geographical and institutional contexts engage in self-directed English learning outside classrooms predominantly through digital platforms, with YouTube, WhatsApp, and social media emerging as the most frequently used tools. A consistent finding is that students favor receptive, input-oriented activities - particularly listening and watching English-language content - over productive activities such as speaking and writing, even when speaking is the skill they most wish to improve. Students typically spend one to three hours daily on English-language digital content, and their self-regulatory practices range from structured goal-setting and cyclical self-reflection to passive engagement such as watching videos without explicit learning strategies. Higher-proficiency learners tend to demonstrate more sophisticated metacognitive awareness and more effective self-regulatory cycles than lower-proficiency peers. The effectiveness of out-of-class self-directed practices depends on the interaction of several factors: individual interest and motivation, self-regulation capacity, degree of institutional scaffolding, and cultural context. Structured programs that bridge classroom instruction with independent learning and include teacher advisory roles produce more systematic and sustained self-directed behaviors than purely informal approaches. Key challenges include time constraints, declining engagement over time, limited productive skill practice, and cultural barriers to online communication in some contexts. Reported language gains center on listening, vocabulary, and speaking confidence, while writing remains the least developed skill through self-directed digital learning. This systematic review of 25 empirical studies underscores the need for targeted scaffolding to enhance the quality and persistence of undergraduate self-directed English learning beyond formal classrooms.

This study investigated the effectiveness of story mps in improving students’ reading comprehension of narrrative texts at the vocational high school level. Many vocational students encounter difficulties in understanding narrative texts, particularly in identifying story elements such as characters, setting, plot, and resolution. Story maps were implemented as a visual learning strategy to assist students in organizing narrative information systematically. This research employed a quantitative approach using a quasi-experimental design. The participants consisted of two classes on eleventh-grade on vocational high school, divided into an experimental group and a control group. The experimental group received instruction using story maps, while the control group was taught using conventional reading methods. Data were collected through reading comehension tests administered as pre-test and post-test. The data were analyzed using normality testing, homogeneity testing, the Mann-Whitney U test, and effect size analysis. The result revealed that there weas no statistically significant difference between the experimental and control groups (p=0.279 > 0.05). However, the experimental group achieved a higher mean score than the control group, indicating a positive effect of story maps on students’ reading comprehension. Although the effect was not statistically significant, the findings suggest that story maps can be an effective instructional strategy to support students’ understanding of narrative texts. Therefore, integrating story maps with other reading strategies is recommended to enhance reading comprehension in vocational school contexts.

Fostering critical thinking competence in nurses: A systematic review.

Transformers have shown remarkable performance in 3D medical image segmentation, but their high computational requirements and need for large amounts of labeled data limit their applicability. To address these challenges, we consider two crucial aspects: model efficiency and data efficiency. Specifically, we propose Light-UNETR, a lightweight transformer designed to achieve model efficiency. Light-UNETR features a Lightweight Dimension Reductive Attention (LIDR) module, which reduces spatial and channel dimensions while capturing both global and local features via multi-branch attention. Additionally, we introduce a Compact Gated Linear Unit (CGLU) to selectively control channel interaction with minimal parameters. Furthermore, we introduce a Contextual Synergic Enhancement (CSE) learning strategy, which aims to boost the data efficiency of Transformers. It first leverages the extrinsic contextual information to support the learning of unlabeled data with Attention-Guided Replacement, then applies Spatial Masking Consistency that utilizes intrinsic contextual information to enhance the spatial context reasoning for unlabeled data. Extensive experiments on various benchmarks demonstrate the superiority of our approach in both performance and efficiency. For example, with only 10% labeled data on the Left Atrial Segmentation dataset, our method surpasses BCP by 1.43% Jaccard while drastically reducing the FLOPs by 90.8% and parameters by 85.8%.

The roles of cognitive and metacognitive strategies in game-based virtual reality learning

This article proposes a unified suboptimal controller design method for unknown general nonlinear systems subject to multiple constraints, including state, input, and output constraints. All inequality constraints are transformed into equality constraints using slack functions and Pade approximation. An unconstrained augmented system is then defined to describe the dynamics of original system and the equality constraints, where the optimal controller of the augmented system can be viewed as a suboptimal controller for the original system. Furthermore, considering the unmodeled dynamics of the original system, neural networks (NNs) are utilized and a data-based solution strategy of integral reinforcement learning (IRL) is presented for the augmented system. Ultimately, the simulation results are given to reflect the effectiveness of the unified design method.