Venue shortages and crowded environments impair college physical education quality, and existing virtual reality (VR) applications lack systematic designs for multi-person synchronous training in confined spaces. This study constructed a multi-loop, VR, confined space training system with a classroom state machine for dynamic regulation and defined the training intensity index and training density for effect evaluation. A randomized, controlled experiment was conducted with 60 college students split into a VR experimental group and a conventional teaching control group for an eight-week intervention. The results showed the VR group had significantly better physical fitness, motor skills, training intensity, training density, and classroom engagement, and it had lower subjective crowding perception. The system achieves high training density without excessive crowding in limited venues, providing a reusable framework and quantitative basis for VR application in college physical education.
As wireless communication technologies advance rapidly, more and more devices are connected via the Internet of Things. This gives rise to numerous IoT applications that generate large volumes of data. Processing these data not only demands computing resources but also consumes significant energy. The emergence of mobile edge computing (MEC) offers a promising paradigm for overcoming the weaknesses of traditional public clouds. As MEC servers are much closer to Internet of Things devices than the public cloud, the transmission delay can be significantly reduced. Nevertheless, the deployment of MEC systems still faces several challenges, including resource allocation and offloading decisions. This paper proposes a joint resource allocation and partial task offloading simulation model for MEC systems that minimizes system costs, including processing time and energy consumption. A mathematical optimization problem is formulated, and a solution algorithm is presented. The simulation results are offered to show the effectiveness of the proposed solution algorithm.
Venue shortages and crowded environments impair college physical education quality, and existing virtual reality (VR) applications lack systematic designs for multi-person synchronous training in confined spaces. This study constructed a multi-loop, VR, confined space training system with a classroom state machine for dynamic regulation and defined the training intensity index and training density for effect evaluation. A randomized, controlled experiment was conducted with 60 college students split into a VR experimental group and a conventional teaching control group for an eight-week intervention. The results showed the VR group had significantly better physical fitness, motor skills, training intensity, training density, and classroom engagement, and it had lower subjective crowding perception. The system achieves high training density without excessive crowding in limited venues, providing a reusable framework and quantitative basis for VR application in college physical education.
During robotic fruit-harvesting operations, the rapid acceleration and deceleration of the robotic arm’s end-effector induce significant shaking of the fruit. These oscillations frequently result in fruit drop, leading to harvesting failures. In this study, tomato fruit bunches were modeled as a rigid–flexible coupling structure, where the stem is treated as a flexible link and the fruit as a rigid payload. An adaptive input-shaping algorithm suppressed oscillations effectively. The amplitude and time of the pulses were dynamically adjusted in real time through an optimized quadratic cost function. This adaptive input-shaping algorithm was simulated on a 3-Degrees of Freedom (3-DOF) harvesting robot. Experimental results demonstrated that this algorithm not only effectively mitigates fruit oscillations but also significantly improves the overall picking success rate.
With the ongoing advancement of digital transformation, cultural heritage dissemination is evolving from being physical preservation to being digital perpetuation. Generation Z, as the core force in digital cultural engagement, relies on immersive experiences, social interaction, and identity formation. Using the Palace Museum's "Cloud Tour of the Palace Museum" as a case study, this research develops a multidimensional UX framework grounded in affordance theory and the S-O-R model to examine how UX shapes eWOM through emotional value and cultural identity. Analysis of 342 survey responses reveals that information and interaction experiences significantly enhance emotional engagement and cultural belonging, while cultural symbol presentation and narrative appeal strengthen cultural identity. This study innovatively applies affordance theory to UX analysis in digital heritage, extends the S-O-R model in digital cultural communication, and provides actionable insights for platform design and sustainable heritage dissemination.
This study investigates the integration of artificial intelligence (AI) and immersive simulation in interactive digital media art by proposing a user-driven framework for cultural heritage education. The framework combined immersive virtual environments with AI-based behavior analysis and adaptive interaction to support personalized user experiences and dynamic content presentation. A simulation-based digital media art prototype was developed and evaluated according to measures of user experience, presence, and learning engagement. Results show that the proposed approach enhances interactivity, immersion, and educational effectiveness compared with traditional digital exhibition methods. The study highlights the potential of AI-enhanced immersive simulation as an effective medium for cultural heritage education and interactive digital media art.
This study explores digital narrative as a computer-mediated simulation to enhance in-service teacher reflection. On the basis of the double-loop reflection-narrative generation model, a toolchain was designed to support script scaffolding, multimodal documentation, and peer mirroring. Over 12 weeks, 54 middle school teachers participated in a mixed-methods experiment. Results showed significantly deeper and more frequent reflections, with weekly reflection frequency rising from 1.4 to 2.6 times (p < 0.01). Qualitative analysis revealed that emotional externalization, peer mirroring, and context replay accounted for 67.3% of reflection initiation. Tool proficiency and self-efficacy significantly moderated reflection depth (beta = 0.42). Findings confirm digital narrative as an effective simulation-based approach for fostering metacognition, emotional expression, and pedagogical growth in teacher training.
Aiming at the problems of high cost, high risk, and difficult to reproduce in traditional coal mine flood rescue drills, this paper proposes a comprehensive drill system based on virtual reality (VR) technology. The system combines three-dimensional scene reconstruction, disaster dynamic simulation, and human-computer interaction design to construct a highly immersive virtual exercise environment. Through the VR universal treadmill, the synchronization of the participants' actions and virtual characters is realized using VR technology to simulate the water inrush process with high fidelity, dynamically present the water flow speed and submerged range, and support multi-role real-time collaborative drills. The supporting evaluation system includes operation scoring, process backtracking, and error analysis, so as to realize quantitative evaluation and continuous optimization of training effect.
With the ongoing advancement of digital transformation, cultural heritage dissemination is evolving from being physical preservation to being digital perpetuation. Generation Z, as the core force in digital cultural engagement, relies on immersive experiences, social interaction, and identity formation. Using the Palace Museum’s “Cloud Tour of the Palace Museum” as a case study, this research develops a multidimensional UX framework grounded in affordance theory and the S-O-R model to examine how UX shapes eWOM through emotional value and cultural identity. Analysis of 342 survey responses reveals that information and interaction experiences significantly enhance emotional engagement and cultural belonging, while cultural symbol presentation and narrative appeal strengthen cultural identity. This study innovatively applies affordance theory to UX analysis in digital heritage, extends the S-O-R model in digital cultural communication, and provides actionable insights for platform design and sustainable heritage dissemination.
This study explored the application of internet of things technology in interactive artmaking as a way to develop interdisciplinary methods for improving the quality of artistic creation. The study integrated theoretical analysis, model construction, and experimental verification to address challenges such as system stability, cross-platform collaboration, and balancing technology with artistry. The research revealed that internet of things integration optimized interactive art by making it more dynamic and responsive to audience behavior-enhancing both artistic language and conceptual depth. The findings contributed valuable insights to the practical optimization of interactive art and emphasized its importance in improving art education and cultural communication. Through investigating the application of internet of things technology in interactive artmaking, this study provides a sustainable framework for future interactive design.
This study addressed the form-function gap in new energy vehicle bionic design by developing a cross-scale framework integrating biological features with engineering needs. Using reverse engineering and dynamic capture, the study created a multidimensional model combining macro-micro structures and dynamic responses, achieving 15.6% aerodynamic improvement through sharkskin/bird bone optimization. Adaptive mesh and zero-order hold methods reduced density errors below 5%. A JavaScript-Object-Notation-based virtual-reality teaching platform enhanced design efficiency and student training. Validated in automotive applications, this work provides quantifiable bionic design tools. Future directions include multimodal sensing, expanded biological prototypes, and deeper materials-bionics integration.
In this study, the authors explore cultural and regional influences on Steam user reviews, highlighting differences in satisfaction between East Asian and Western players. Using a multimethod approach-statistical testing, clustering, and sentiment analysis with large language models on a curated Steam review dataset-the authors found that East Asian users (especially from China, Japan, and South Korea) consistently rate games lower than the global average, with Simplified Chinese reviewers giving the lowest scores worldwide. In contrast, English-speaking users align closely with the global mean. Sentiment analysis revealed that Simplified Chinese negative reviews express more anger, whereas their positive reviews show less joy compared with English ones. These patterns likely reflect cultural differences in expectations about game quality, price sensitivity, and aesthetic preferences. The results emphasize how cultural context shapes player satisfaction and offer actionable insights for developers to improve user experience through culturally aware design.
This study presents a framework integrating artificial intelligence, augmented reality, and neurophysiological assessment for preserving and digitizing Chinese Keju cultural heritage. Through the implementation of deep learning algorithms, including convolutional neural network with long short-term memory, Transformer with optical character recognition, YOLO-v8 with bidirectional encoder representations from Transformers, and GPT-4V with augmented reality, the authors achieved progressive accuracy improvements from 0.70 to 0.98 across 100 training epochs for cultural artifact recognition tasks. The 32-channel electroencephalogram experiments (n = 25) comparing physical and virtual reality product appreciation revealed distinct neural activation patterns, with physical interaction eliciting improved multisensory integration through enhanced parietal alpha (8-13 Hz) and sensorimotor cortex activation, while virtual reality demonstrated dominant occipital gamma activity, indicating visual processing emphasis. The augmented reality processing pipeline achieved 93.4% throughput with 85.9-ms latency, enabling real-time cultural heritage experiences.
This study explored the application of internet of things technology in interactive artmaking as a way to develop interdisciplinary methods for improving the quality of artistic creation. The study integrated theoretical analysis, model construction, and experimental verification to address challenges such as system stability, cross-platform collaboration, and balancing technology with artistry. The research revealed that internet of things integration optimized interactive art by making it more dynamic and responsive to audience behavior—enhancing both artistic language and conceptual depth. The findings contributed valuable insights to the practical optimization of interactive art and emphasized its importance in improving art education and cultural communication. Through investigating the application of internet of things technology in interactive artmaking, this study provides a sustainable framework for future interactive design.
As digital media art education increasingly relies on simulation-based environments such as virtual reality and interactive design platforms, the need for adaptive learning support grows. Traditional approaches often fail to meet students’ diverse and evolving creative needs. This study presents an AI-driven personalized recommendation system that uses multimodal data fusion and deep learning to model learners’ behavior, aesthetic preferences, and creative development in digital art simulations. An eight-week experiment with university students shows the system significantly improves recommendation accuracy (88.7%), user satisfaction (90.5%), and content diversity over conventional methods. Results indicate that the system enhances creative engagement and supports adaptive learning, offering an effective AI-powered solution for intelligent support in simulation-based art education.
This study addressed the form-function gap in new energy vehicle bionic design by developing a cross-scale framework integrating biological features with engineering needs. Using reverse engineering and dynamic capture, the study created a multidimensional model combining macro-micro structures and dynamic responses, achieving 15.6% aerodynamic improvement through sharkskin/bird bone optimization. Adaptive mesh and zero-order hold methods reduced density errors below 5%. A JavaScript-Object-Notation-based virtual-reality teaching platform enhanced design efficiency and student training. Validated in automotive applications, this work provides quantifiable bionic design tools. Future directions include multimodal sensing, expanded biological prototypes, and deeper materials-bionics integration.
This study presents a simulation-driven framework that transforms marine organism morphologies into culturally meaningful design concepts through parametric modeling, semantic mapping, and user-perception feedback. By integrating digital extraction, generative design, and iterative evaluation, the method bridges biological inspiration with cultural expression. Results show a 93.0% contour accuracy (IoU) in seahorse modeling, a cultural recognition score of 4.4/5 after three design iterations, and a visual appeal score of 8.9/10 for jellyfish-inspired designs. The framework employs computer-mediated simulation to visualize and refine bionic forms in real-time design environments. User perception data are fed back into the generative loop, enabling adaptive optimization of aesthetic and cultural coherence. The approach demonstrates how simulation-based design can enhance functional, emotional, and cultural integration in creative applications, with potential implications for serious games, virtual exhibitions, and interactive storytelling.
In this study, the authors explore cultural and regional influences on Steam user reviews, highlighting differences in satisfaction between East Asian and Western players. Using a multimethod approach—statistical testing, clustering, and sentiment analysis with large language models on a curated Steam review dataset—the authors found that East Asian users (especially from China, Japan, and South Korea) consistently rate games lower than the global average, with Simplified Chinese reviewers giving the lowest scores worldwide. In contrast, English-speaking users align closely with the global mean. Sentiment analysis revealed that Simplified Chinese negative reviews express more anger, whereas their positive reviews show less joy compared with English ones. These patterns likely reflect cultural differences in expectations about game quality, price sensitivity, and aesthetic preferences. The results emphasize how cultural context shapes player satisfaction and offer actionable insights for developers to improve user experience through culturally aware design.
This study investigates Japanese parents’ perceptions of the educational value of esports and how their mediation strategies shape these views. A nationwide online survey examined parental expectations across non-cognitive, personality, and interpersonal domains, alongside three mediation styles: restrictive, active, and co-playing. Structural equation modeling identified “esports readiness”—driven by parental involvement in education, personal gaming habits, and esports literacy—as a key predictor of mediation practices. Active mediation and co-playing were positively associated with perceived educational benefits, while restrictive mediation had a negative effect. Findings highlight the need for balanced engagement strategies and parent-inclusive design in educational esports programs. This work offers evidence-based guidance for educators, policymakers, and designers aiming to integrate esports into youth skill development in culturally sensitive ways.
To address the challenges of intensive task demands, limited resources, and strong operational coupling at dry bulk terminals, this article proposes a Differential Learning-Based Memetic Algorithm (DL-MA) for integrated scheduling of unloading and maintenance tasks. The algorithm embeds a differential learning mechanism within a global evolutionary framework, combining feasibility repair and mild perturbation to guide solutions toward feasible and high-quality regions. A multi-dimensional scoring function is constructed to assess task priority satisfaction, resource balance, and robustness. Experiments under varying task loads and capacity constraints demonstrate that DL-MA consistently outperforms a benchmark genetic algorithm in task completion, fitness value, and structural diversity. The problem addressed is central to intelligent port operations and has broad application potential. The proposed approach is also scalable and adaptable to other complex scheduling environments, offering a robust and efficient tool for intelligent production planning.
