Search Results (5,895)

In the field of forensic document examination, accurately determining the chronological sequence of intersecting lines between seal ink and handwriting is a crucial technical step for verifying document authenticity, identifying contract tampering, and detecting forged signatures. This technique analyzes the physical superimposition relationship formed by the deposition of the two media on the paper substrate to provide objective scientific evidence for judicial practice. Although traditional methods such as microscopic imaging and mass spectrometry analysis have achieved some progress, they still suffer from common limitations including high equipment costs, complex operation, and potential damage to samples. This study proposes and validates an innovative non-destructive determination method that integrates structured light 3D reconstruction technology with deep learning algorithms. The research captures the microscopic 3D morphological features of the ink intersection area using a high-precision structured light scanning system and effectively eliminates noise interference caused by paper substrate undulation through Gaussian flattening technology. Subsequently, a multimodal fusion strategy combines 2D texture images with 3D depth information to construct a dataset rich in features. On this basis, a deep learning model based on an improved Residual Neural Network (ResNet) is designed, incorporating the ELU activation function and an EMA mechanism to enhance the model’s feature extraction capability and convergence stability. Experimental results demonstrate that the proposed method achieves a recognition accuracy of 94.39% on the test set, fully validating its effectiveness and application potential in the non-destructive determination of ink stroke sequencing. Full article

Against the backdrop of China’s rural revitalization, understanding the factors influencing farmers’ agricultural technology adoption behavior is crucial for enhancing such adoption. Therefore, exploring the decision-making logic behind farmers’ agricultural technology adoption behavior is of paramount importance. This study, conducted among 482 typical farming households in the Chengdu Plain of Sichuan Province, China, introduced the Random Forest (RF) algorithm into an Improved TAM. Combined with SHAP and PDP techniques, it identified 21 influencing factors and their nonlinear interaction mechanisms. Key findings include the following: (1) Adoption rates stood at only 14.3%, exhibiting a pronounced “advantage-oriented” pattern favoring male farmers, middle-aged/young adults, and higher-income groups; (2) Level of agricultural production tools and technology (C1) and Agricultural product sales channels (C2) emerged as core drivers, with C1 presenting a significant “technology threshold effect”—adoption probability surged from 0.1 to over 0.35 during intelligent technology transitions; (3) Monthly household income level (B4) effectively mitigates risk aversion among elderly farmers, revealing the critical role of Age (A2) in decision-making and enabling a complementary relationship between experience and technology; (4) Self-learning and training proficiency in agricultural technology (F1) reflects that excessive technological complexity triggers resistance and blocks adoption, while Highest educational attainment in the household (B1) and Number of educated family members (B2) exhibit nonlinear peak characteristics influenced by “brain drain” due to labor migration. These findings not only expand the theoretical application of machine learning in studying farmer behavior but also provide granular insights for overcoming the “last mile” bottleneck in agricultural technology dissemination. Full article

Rapid advancements in artificial intelligence (AI) technologies, coupled with UNESCO’s Education 2030 vision, necessitate a re-evaluation of teachers’ technological and pedagogical competencies aligned with sustainability goals. This study investigates the impact of pre-service teachers developing digital materials within the framework of the Sustainable Development Goals (SDGs) using AI and AI-supported Web 2.0 tools (e.g., ChatGPT, DeepSeek, Alayna, Padlet, Canva, Kahoot) on their Artificial Intelligence Technological Pedagogical Content Knowledge (AI-TPACK) levels. Employing an explanatory sequential mixed-methods design, the research was conducted with 31 pre-service teachers over a 10-week applied training period. Data were collected via the AI-TPACK Scale and semi-structured interviews. Quantitative findings revealed that the applied training significantly enhanced the pre-service teachers’ Pedagogical Knowledge (PK), AI-Technological Knowledge (AI-TK), Pedagogical Content Knowledge (PCK), and overall AI-TPACK levels. However, no statistically significant difference was observed in the Content Knowledge (CK) dimension. Qualitative data demonstrated that AI-supported tools made the learning environment more engaging and efficient, concretized abstract sustainability concepts, and bolstered the pre-service teachers’ digital self-confidence. Consequently, this study establishes that integrating AI tools into SDG education is an effective strategy for cultivating pre-service teachers’ technopedagogical competencies, empowering them to perceive technology as a facilitator of professional development rather than an instructional barrier. Full article

Cultural heritage, in all its tangible and intangible expressions, is undergoing a process of renewal driven by the integration of digital technologies and participatory approaches. This study presents a pilot project developed within the SAMOTHRACE Fundation, focused on the design of a Serious Game dedicated to the Norman Castle of Aci Castello in Sicily. The project explores how game-based learning and interactive storytelling can enhance visitor engagement, accessibility, and understanding of small-scale heritage sites that are often excluded from major cultural circuits. Using Unity and Blender, the prototype combines historical research, 3D reconstruction, and narrative interaction to transform the castle into an immersive educational environment. This initial phase also served as the basis for an academic thesis, laying the methodological groundwork for future expansion and evaluation. The results of this pilot provide preliminary quantitative evidence that serious games can support cultural communication strategies, foster emotional engagement, and stimulate curiosity toward minor heritage sites, while remaining compatible with the constraints of modest institutions. Ultimately, the project illustrates how even modest institutions can leverage digital innovation to revitalize their heritage assets, promote inclusive participation, and explore new models of interactive archaeology and community-centered cultural engagement. Full article

The combination of Artificial Intelligence (AI) and Formative Assessment (FA) in Teacher Education explores how emerging technologies can enhance teaching practices and professional development. AI tools can provide personalized feedback, identify learning needs, and support reflective practice among educators. Integrating AI-driven formative assessment methods allows for continuous evaluation of teaching competencies, promoting adaptive learning, data-informed decision-making, and improved instructional quality in teacher education programs. The purpose of this study was to conduct a systematic review of the use of Formative Assessment (FA) and Artificial Intelligence (AI) in Teacher Education (TE) during the period 2020–2025 (inclusive). The review was conducted following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) methodology, which ensures a rigorous, transparent, and reproducible process in the selection and analysis of studies. To this end, scientific articles published in the Scopus, Web of Science and Dialnet databases were reviewed, considering publications in English and Spanish. The objective was to identify trends, methodological approaches, results, and research gaps that show how AI is being integrated, or not, into FA processes in TE. The review also sought to analyze the impact of AI on student participation in assessment, feedback, decision-making, and the learning and assessment process itself, synthesizing the current evidence on the relationship between AI and FA in TE. Full article

Artificial intelligence (AI) technologies are becoming increasingly integrated into the everyday lives of children, influencing how they learn, communicate, and develop emotionally. As AI assumes a more central role in children’s digital ecosystems, AI literacy—the ability to understand, engage with, and make informed decisions about AI systems—is no longer a luxury but a developmental necessity. This review explores how AI literacy intersects with children’s mental health, particularly through the lens of socioeconomic status. Drawing on Digital Capital Theory and Cumulative Advantage/Disadvantage Theory, the paper examines how inequalities in access to AI-related resources shape the emotional and psychological experiences of children. It argues that while AI literacy can enhance well-being across all social groups, its impact is especially transformative for children from low-income backgrounds. Children from middle- and high-income families often experience modest emotional gains from AI engagement, having already benefited from consistent digital exposure and support. In contrast, low-income children—who often begin with limited access and lower confidence—stand to gain disproportionately in terms of emotional resilience, self-esteem, and digital confidence when their AI literacy improves. The review concludes with policy and practice recommendations that prioritize equitable access and tailored interventions, especially for underserved populations who have the most to gain from both the cognitive and emotional benefits of AI literacy. Full article

With the development of 6G communication technology, Space–Air–Ground Integrated Networks (SAGINs) have become critical infrastructure for global intelligent collaborative computing. However, federated learning deployment in SAGINs faces three severe challenges: the high dynamics of satellite orbital motion, node resource heterogeneity, and privacy vulnerabilities in data transmission. This paper proposes an adaptive compressed sensing differential privacy federated learning framework based on orbital spatiotemporal characteristics. First, we design orbital periodicity-driven time-varying sparse sensing matrices that dynamically adjust compression strategies according to satellite orbital positions, achieving intelligent communication efficiency optimization. Second, we propose an orbital predictability-based privacy budget temporal allocation mechanism and perform differential privacy noise injection in the compressed domain, establishing a compression–privacy joint optimization algorithm. Furthermore, we construct an energy–communication–privacy ternary collaborative mechanism that achieves multi-objective dynamic balance through model predictive control. Finally, we design reinforcement learning-based dynamic routing scheduling and hierarchical aggregation strategies to effectively handle the time-varying characteristics of network topology. Simulation experiments demonstrate that compared to existing methods, the proposed approach achieves 3–12% improvement in model accuracy and 30–50% enhancement in communication efficiency while maintaining differential privacy protection with dynamic privacy budget ε ∈ [0.1,10.0]  and compression ratio ρ ∈ [0.2,0.8]. Unlike static compressed sensing approaches that ignore orbital periodicity, the proposed orbital-driven time-varying sensing matrices reduce reconstruction error by up to 19.4% compared to fixed-matrix baselines, validating the synergistic effectiveness of integrating orbital spatiotemporal characteristics with federated learning in 6G SAGIN deployments. The framework assumes reliable orbital propagation via SGP4/SDP4 models and does not account for Doppler frequency shifts or inter-satellite link handover delays; future extensions include scalability to mega-constellations and integration of quantum-resistant privacy mechanisms. Full article

The future of innovation and economic growth depends on our ability to nurture the next generation of scientists. The global shortage of qualified STEM (Science, Technology, engineering, Mathematics) teachers has led many countries to expedite the transition of subject-matter experts from industry and academia into teaching roles. These second-career science student teachers typically participate in accelerated training programs designed to address urgent shortages. This study addresses a gap in the literature regarding effective pedagogical interventions for career-changing professionals in STEM fields, focusing on the experience and transformation of second-career science student teachers. This qualitative case study explores how a Challenge-Based Learning (CBL) course fosters the development of pedagogical competences via developing an instructional unit collaboratively, among five second-career science student teachers enrolled in an accelerated teacher education program. Drawing on data collected through instructors’ field notes, iterative work-in-progress lesson drafts, and reflective final papers, the study employs qualitative content analysis to trace changes in participants’ instructional approaches and professional identity. Findings reveal that engagement with the CBL framework promoted a significant shift from teacher-centered to learner-centered instruction, as participants increasingly integrated collaborative learning, inquiry-based activities, and reflective practices into their lesson planning and classroom teaching. The iterative nature of CBL, which emphasizes real-world problem-solving and structured opportunities for reflection and peer feedback, was instrumental in supporting participants’ adaptive expertise and confidence as novice teachers. Moreover, the course experience contributed to the emergence of a professional teaching identity, with participants reporting greater self-efficacy, a stronger sense of belonging to the teaching community, and increased motivation to persist in the profession. The results underscore the potential of integrating CBL and learning sciences principles into accelerated teacher preparation programs to enhance both cognitive and affective dimensions of teacher development. Full article

Integration of artificial intelligence (AI) into mathematics education holds significant potential to enhance learning outcomes; however, its effectiveness in resource-constrained higher education contexts remains uneven due to persistent digital divide barriers. This quantitative study investigates how socioeconomic status shapes first-level (technology access) and second level (digital skills and institutional support) digital divide barriers, and how these factors relate to students’ perceptions of AI-driven mathematics learning. Grounded in van Dijk’s digital divide theory, a cross-sectional survey was administered to 121 undergraduate mathematics students at a historically disadvantaged higher education institution. Descriptive statistics, Pearson correlation, and Chi-square analyses were employed to examine associations among socioeconomic status, access, skills, institutional support, and AI perceptions. The findings indicate that material access barriers, such as limited devices and internet connectivity, remain prevalent among disadvantaged students but show weak or inconsistent associations with perceptions of AI. In contrast, institutional support demonstrates a statistically significant positive relationship with students’ perceptions of AI training (r = 0.212, p < 0.05), highlighting its central role in shaping readiness for AI-enhanced learning. Overall, the results suggest that second-level digital divide factors, particularly structured institutional support, are more influential than access alone in determining students’ engagement with AI in mathematics education. The study implies the need for universities to move beyond infrastructure provision toward comprehensive and sustained institutional strategies that foster confidence, guided engagement, and equitable AI adoption. Full article

As network slicing gains traction in cloud computing environments, efficient management and orchestration systems are required to realize the benefits of this technology. These systems must enable dynamic provisioning and resource optimization of virtualized services spanning multiple network slices. Nevertheless, the common resource overprovisioning practice implemented by service providers leads to the inefficient use of resources, limiting the ability of Mobile Network Operators (MNOs) to rent new network slices to more vertical customers. Hence, efficient resource allocation mechanisms are essential to achieve optimal network performance and cost-effectiveness. This paper proposes a predictive model for network slice resource optimization based on resource sharing between Virtualized Network Functions (VNFs). The model employs deep learning models based on Long Short-Term Memory (LSTM) and Transformers for CPU resource usage prediction and a reactive algorithm for resource sharing between VNFs. The model is powered by a telemetry system proposed as an extension of the 3GPP network slice management architectural framework. The extended architectural framework enhances the automation and optimization of the network slice lifecycle management. The model is validated through a practical use case, demonstrating the effectiveness of the resource sharing algorithm in preventing VNF overload and predicting resource usage accurately. The findings demonstrate that the sharing mechanism enhances resource optimization and ensures compliance with service level agreements, mitigating service degradation. This work contributes to the efficient management and utilization of network resources in 5G networks and provides a basis for further research in network slice resource optimization. Full article

Rare Earth Elements (REEs) represent strategic and critical raw materials for the energy transition and must therefore be integrated into efficient and functional recycling processes. Their adoption in electric motors is rapidly expanding, raising significant challenges for end-of-life (EoL) management, starting from the collection phase. In this context, this work proposes the integration of an image-based classification framework within the Waste Electrical and Electronic Equipment (WEEE) recycling pipeline to selectively identify electric motors containing permanent magnets (PMs) and direct them toward dedicated recycling processes for rare earth recovery. The proposed methodology relies on a Discriminative Transfer Learning (DTL) approach based on a ResNeXt convolutional neural network (CNN), adapted to a proprietary and heterogeneous dataset of electric motors acquired in an industrial recycling facility. The objective is twofold: first, to identify motors containing PMs; second, to classify motors into construction categories according to their likelihood of incorporating PMs. Experimental results show promising performance in terms of PM-containing motor detection capability, establishing a robust foundation for the automated recovery of REEs at an industrial scale. Furthermore, the model’s generalization capabilities can be further enhanced through the expansion of collaborative datasets and the integration of advanced scanning technologies. Full article

Slow-moving landslides can trigger severe disasters when activated by earthquakes, torrential rains, or typhoons. Early detection is crucial for mitigating loss of life and property damage. Interferometric Synthetic Aperture Radar (InSAR) technology is among the most effective techniques for detecting slow-moving landslides, though its accuracy can be further improved through integration with optical imagery and Digital Elevation Models (DEM). Current machine learning approaches that combine InSAR and optical data suffer from limited efficiency, poor transferability, and challenges in regional-scale application. To address these limitations, this study proposes a multimodal dual-path network that integrates InSAR products with textural information from optical imagery to detect slow-moving landslides. One path processes InSAR deformation rates and topographic factors, while the other incorporates texture information and auxiliary data. Together, these paths extract semantic information from high-dimensional spatial features and condense it into low-dimensional representations. A pyramid pooling module is employed to capture multi-scale features during low-level semantic extraction. For feature fusion, a rate-constrained adaptive module is introduced to enhance the contribution of deformation rates to slow-moving landslides. According to the results, the proposed method improves the F1-score for landslide detection by 6% compared to using InSAR products alone. These results provide reliable technical support for regional landslide inventory compilation and disaster management, as well as new insights for regional-scale surveys in slow-moving landslide-prone areas. Full article

Green technological innovation is a core driving force for firms’ low-carbon transformation. However, because critical green technologies and knowledge are often dispersed across upstream and downstream partners within supply chains, firms’ green transformation faces substantial challenges. Previous studies have primarily focused on internal drivers at the firm level while overlooking the empowering role of green collaborative cooperation among supply chain partners. To address this gap, this study introduces empowerment theory to systematically examine how supply chain green strategic alliances enhance firms’ green innovation capability. Using a sample of Chinese A-share listed firms from 2011 to 2023, we construct a firm-level indicator of supply chain green strategic alliances based on textual analysis and machine learning techniques and empirically test its impact on green innovation. The results show that participation in green strategic alliances significantly promotes firms’ green innovation. Mechanism analyses further reveal that this effect operates through the reconstruction of green knowledge, increased environmental investment, and improved green governance. Moreover, the positive effect is more pronounced in regions with stronger intellectual property protection, greater green credit support, and stricter environmental regulation, as well as among firms with closer supply chain relationships. This study identifies supply chain green strategic alliances as a key inter-organizational empowerment mechanism and provides important practical implications for leveraging supply chain collaboration to accelerate sustainable development and firms’ green transformation. Full article

Quantum computing offers potential computational advantages, yet its integration into autonomous decision-making systems remains largely unexplored. This paper addresses the need for a unified framework that systematically combines quantum computation with agent-based artificial intelligence. We examine how quantum technologies can enhance the capabilities of autonomous agents and, conversely, how agentic AI can support the advancement of quantum systems. We analyze both directions of this synergy and present conceptual and technical foundations for future quantum–agentic platforms. Our work introduces a formal definition of quantum agents and outlines architectures that integrate quantum computing with agent-based systems. As concrete proof-of-concept implementations, we develop and evaluate three quantum agent prototypes: (i) a Grover-based decision agent for quantum search-driven action selection, (ii) a variational quantum reinforcement learning agent for adaptive policy learning in a multi-armed bandit setting, and (iii) an adaptive quantum image encryption agent that autonomously selects encryption strategies based on entropy-driven feedback. These prototypes demonstrate practical realizations of quantum agency in decision-making, learning, and security contexts under NISQ-era constraints. Furthermore, we discuss application domains including quantum-enhanced optimization, hybrid quantum–classical orchestration, autonomous quantum workflow management, and secure quantum information processing. By bridging these fields, we introduce a structured theoretical and architectural framework for quantum–agentic systems, providing formal definitions, system models, and early operational prototypes that illustrate the feasibility of quantum-enhanced agency under NISQ constraints. Full article

Prostate cancer (PCa) remains a major global health challenge, yet conventional diagnostic methods are often limited by suboptimal accuracy and efficiency. Artificial intelligence (AI) has emerged as a rapidly developing technology capable of integrating multi-source data to enhance clinical decision-making. This narrative review synthesizes current evidence regarding AI applications across key diagnostic domains, including medical imaging, digital pathology, liquid biopsy, and multi-omics integration. Findings indicate that AI models for magnetic resonance imaging (MRI) can improve risk stratification and may reduce unnecessary biopsies in some cohorts, particularly when evaluated alongside structured radiology assessment and clinical variables. In digital pathology, deep learning algorithms have shown high agreement with expert genitourinary pathologists for automated Gleason grading in controlled and externally validated settings, with potential to reduce reporting time for high-volume workflows. Additionally, AI-powered liquid biopsy models may support non-invasive risk stratification, particularly for patients with prostate-specific antigen (PSA) levels in the diagnostic gray zone, while multi-omics integration is being investigated to enhance personalized assessment. Despite advances, challenges regarding data heterogeneity, algorithm interpretability, and workflow integration persist. Future research should prioritize multimodal data fusion, explainable AI development, robust calibration and decision-analytic evaluation, and large-scale prospective validation to standardize protocols and fully realize the potential of AI in precision prostate cancer care. Full article