Search Results (2,917)

Several countries have adopted a nationwide ban on social media access for children. Our aim was to investigate public opinion regarding the implementation of a social media ban for children, as well as the factors influencing these views. We measured agreement with the ban, information regarding its implementation, perceived need for additional measures, confidence in the effectiveness of the ban, perceived impact of the ban, and parental familiarity with digital parental control tools. The study sample included 619 participants. In our sample, 69% agreed with the implementation of the ban, while 86.5% believed that additional measures should be implemented (i.e., digital literacy courses in schools, active parental involvement in digital literacy, prohibition of inappropriate content, reasonable parental limits on social media use, and restriction of addictive platform features). Females and higher-educated participants had more positive perceptions regarding the impact of the ban. We found a positive association between age, financial status, social media use, and impact of the ban. Reduced age was associated with increased parental familiarity with digital parental control tools. Social media use was associated with parental familiarity with digital parental control tools. There is a need for holistic and evidence-informed policy frameworks that integrate regulatory measures, educational initiatives, and shared accountability among stakeholders. Full article

Higher education policy on generative artificial intelligence has developed rapidly, yet much of this development remains stronger on governance, permission, disclosure, and assurance than on pedagogy. Universities increasingly move beyond blanket prohibition by distinguishing between restricted and permitted contexts, requiring acknowledgement of tool use, and introducing verification mechanisms to protect authorship and understanding. However, publicly visible institutional approaches appear less developed in providing structured, student-facing workflows that guide responsible AI engagement during assessment completion. This article, informed by a bounded qualitative document analysis, uses the term pedagogical middle layer to describe the process guidance needed between institutional permission settings and academic-integrity or misconduct procedures. Drawing on recent literature and a purposive scan of selected publicly available university policy and guidance documents, the paper argues that current public-facing models are often effective at defining boundaries but less explicit in guiding disciplined, transparent, and defensible forms of human–AI collaboration. In response, the paper presents the Structured AI-Guided Education Assessment Policy (SAGE-AP) as a theoretically grounded policy proposal for AI-assisted assessment, rather than as an empirically validated policy intervention. SAGE-AP frames assessment as a staged process in which students begin from their own understanding, engage with AI critically, document evaluative decisions, refine outputs responsibly, and defend the reasoning represented in the final submission. The paper contributes to institutional policy development by clarifying how permission settings may be complemented by pedagogical process guidance in the generative AI era. Full article

As the environmental sustainability of blockchain technology becomes a focal point of public and academic debate, understanding how technically engaged communities frame this issue is increasingly important. This study examines 3000 long-form comments from a highly active sustainability-focused Bitcointalk thread to analyze sentiment patterns, recurring arguments, and the linguistic cues associated with community responses to environmental criticism. Using Natural Language Processing (NLP) methods, we apply Valence Aware Dictionary and sEntiment Reasoner (VADER) sentiment analysis to classify the discourse, n-gram extraction to identify dominant thematic expressions, and a Random Forest model combined with SHapley Additive exPlanations (SHAP) to interpret the lexical features most strongly associated with sentiment polarity. The results show a strongly positive and internally consistent discourse structure: 87.63% of comments are classified as positive, while negative and neutral comments are comparatively rare. The dominant themes emphasize energy consumption as a necessary trade-off for network security, while external criticism is frequently reframed or rejected. Explanatory modeling further indicates that negative sentiment is primarily driven by terms associated with climate risk, damage, and reputational concerns when users respond to criticism. Rather than claiming to capture the cryptocurrency ecosystem as a whole, this study presents a localized case study of one Bitcointalk mega-thread and describes it as a highly homogeneous narrative space shaped by recurrent rebuttal and rhetorical reinforcement. The findings offer a focused contribution to understanding how insider communities construct sustainability narratives around blockchain energy use, while also highlighting the need for broader comparative and network-structural research in future work. Full article

Low-carbon operation and reasonable carbon responsibility allocation are essential for improving source-load coordinated emission reduction in park integrated energy systems (PIESs). Existing allocation methods usually trace carbon emissions or calculate marginal contributions, but they still have difficulty distinguishing heterogeneous park users with different load rigidity, demand response (DR) capability, payment capability and real carbon-reduction potential. To address this problem, this paper proposes a carbon responsibility allocation method for PIESs considering user heterogeneity and develops a carbon-cost-feedback-based bi-level low-carbon scheduling model. First, park users are classified into high-energy-consuming industrial users, commercial and public service users, and energy infrastructure users according to quantitative criteria related to energy consumption scale, load continuity, adjustable load proportion and distributed-resource interaction capability. A heterogeneity indicator system is then established, including DR elasticity, electricity utilization efficiency, payment capability, DR potential and actual carbon-reduction potential. Second, an improved Shapley value allocation model is constructed by combining coalition marginal contribution with entropy-weighted heterogeneity correction. The allocation results are converted into user-side carbon responsibility cost signals and embedded into a bi-level optimal scheduling model, where the upper level minimizes the system operating cost and the lower level minimizes users’ integrated energy-use cost. Case studies show that, compared with the conventional economic scheduling scenario, the proposed model reduces the total system cost from CNY 5.0782 million to CNY 4.3258 million and decreases carbon emissions from 14,994.39 t to 10,874.62 t, corresponding to reductions of 14.82% and 27.47%, respectively. The results indicate that the proposed method can coordinate fairness-oriented carbon responsibility allocation with incentive-oriented low-carbon scheduling, supporting both SDG 11 and SDG 12. Full article

The study aimed to examine the role of artificial intelligence in preservice science teachers’ analogical reasoning by comparing the features of analogy designs produced with and without artificial intelligence. The research was conducted with 133 preservice science teachers at a public university in Türkiye. Participants were divided into two conditions: those who designed analogies using artificial intelligence (n = 62) and those who designed analogies without artificial intelligence (n = 71). Analogy design products were analyzed using descriptive analysis, and categorical data derived from these analyses were examined through Pearson’s chi-square tests. In addition, qualitative data obtained from structured interviews with the AI-supported condition were analyzed using content analysis. The results revealed significant differences between the groups in several dimensions of analogy design, presentation format, semantic distance, analogical association, wealth level, and the identification of limitations. Analogies designed with artificial intelligence were more frequently pictorial–verbal, involved both close and remote semantic distance, integrated structural–functional associations, and exhibited extended analogy characteristics. Interview results indicated that preservice science teachers primarily used AI for idea generation, visualization, and creative exploration rather than for generating factual knowledge. These results contribute to the literature by highlighting the potential role of AI in supporting representational transformation processes within science teacher education. Full article

Research Objective: To explore governance pathways for small-property-right housing from the perspective of land appreciation revenue distribution, thereby promoting high-quality urban renewal. Research Methods: The study employs theoretical analysis, inductive summarization, and logical reasoning. Research Findings: (1) Land appreciation revenue consists of absolute rent, differential rent I, and differential rent II, corresponding respectively to land ownership, land development rights, and land management rights; (2) A framework for the distribution of land appreciation gains that “balances public and private interests and promotes multi-stakeholder sharing” is established, clarifying the revenue boundaries for entities such as the government, village collectives, and housing operators; (3) Two governance pathways are proposed: converting retained collective property rights into affordable rental housing, and categorizing and disposing of properties after government expropriation and conversion to state ownership. These are further refined into five implementation models. Research Conclusions: The rational distribution of land appreciation gains is key to resolving the governance challenges of small-property-right housing and coordinating the objectives of urban renewal with housing security. Full article

Background: The nationally advertised mass media campaign Act-FAST UK, delivered in multiple waves since its launch in 2009, has increased public awareness of stroke symptoms. However, many stroke patients still delay in calling for help and reach the hospital too late to receive emergency treatments. The reasons for this cognitive dissonance between recognition of symptoms and urgent seeking of emergency medical services (EMS) are unclear. Aims: This study aimed to quantify cognitive, psychological, and knowledge-based barriers to help-seeking in patients with acute stroke or transient ischaemic attack (TIA), as well as in intervening witnesses, and to examine their association with the use of EMS as the initial point of contact. Methods: We interviewed patients admitted to a hyperacute stroke unit with a stroke or transient ischaemic attack (TIA) from 2013 to 2016. People who contacted emergency services on the patient’s behalf (intervening witnesses (IWs)) were also interviewed when available. Reasons given for delays in calling for help were related to correct symptom recognition, and whether/at what time, emergency services were contacted after symptoms onset. Results: A total of 602 patients (429 with stroke, 173 with TIA) along with 128 witnesses who intervened in calling for help in those cases (IWs) were interviewed. In the subset of patients with both measures available, there was a strong positive correlation between NIHSS score and number of FAST symptoms (Spearman’s rho = 0.645, p < 0.001), providing supportive evidence for the use of FAST symptom count as a proxy measure of stroke severity. A total of 469 (77.9%) of the patients were aware of a media education campaign about stroke, but only 145 (24.1%) had attributed their own symptoms to stroke at onset. However, correct self-diagnosis of stroke was not associated with direct calls to the EMS (OR 1.43, 95% CI 0.84–2.45). Cognitive, psychological or emotional barriers to help-seeking, as reported by prior published studies, were reported by 463 (81.2%) of the patients we interviewed but in only 63 (53.3%) of the IWs (p < 0.001). Amongst the patient cohort, “not thinking symptoms were serious” (275, 45.7%) and “waiting to see if symptoms would go away” (285, 47.3%) were most strongly negatively associated with EMS use (OR 0.52, 95% CI 0.32–0.84 and OR 0.34, 95% CI 0.21–0.55, respectively). Only 55 (9.1%) of the patients interviewed had been aware of any time-critical stroke treatment prior to their stroke. Eighteen stroke patients (4.2%) reached hospital in time to receive thrombolysis, but an additional 170 (39%) could have been considered for this treatment (i.e., had no apparent other contraindications from a notes review) had they arrived within 4 h of symptom onset. Conclusions: Future public education campaigns may be more effective if they specifically address factors associated with delays in calling for help after stroke symptoms and emphasise the existence of emergency treatments, which are also time-critical. More effective public education may have the potential to increase the proportion of patients arriving in time to benefit from such treatments. Full article

Background: Antimicrobial resistance, driven by inappropriate use and overuse of antibiotics, is a major public health threat. Diarrhea and respiratory illness are the leading causes of pediatric healthcare visits in low- and middle-income countries like Bangladesh. Despite clear WHO guidelines recommending limited use of antibiotics for these conditions, potentially inappropriate or non-prescription antibiotic use remains a concern. Methods: We interviewed caregivers of 3025 under-5 children via cellphones to assess common illnesses, associated care-seeking practices, and antibiotic use for diarrhea and respiratory illnesses experienced by their children in the prior 14 days. Caregivers were identified through hospital outpatient screening and were contacted over the phone for the interview at least two months after that hospital visit. Results: Among the participants, 116 (3.8%) reported diarrheal disease and 570 (18.8%) experienced respiratory illness during the preceding 2-week recall period. Among the children with diarrhea, 52.6% received antibiotics, and 73.8% obtained them over the counter from pharmacies. Among those with respiratory illness, 26.3% received antibiotics, and 58% procured them from local drugstores without a prescription from a registered physician. For diarrhea, azithromycin and metronidazole were the commonly used antibiotics, while for respiratory illness, cefixime and azithromycin were frequently used. Notably, 68% of the diarrheal children either sought care from local drugstores, were self-medicated, or did not receive any formal treatment. Conventional practice, long wait times at healthcare facilities, distance, and poverty were the main reasons for not seeking care from a registered healthcare provider. Conclusions: Understanding community-level antibiotic use and care-seeking behavior is essential to strengthening antibiotic stewardship and child health programs. Our findings suggest the need for context-sensitive community education, improved access to appropriate care, and enforcement of regulations restricting the over-the-counter sale of antibiotics to curb irrational and excessive antibiotic use. Full article

Manual classification of breast cancer is resource-intensive, slow, and subject to inter-observer variability, motivating automated deep learning solutions. Most current methods rely on unimodal imaging data and struggle with domain generalization (DG) across varied clinical environments. We propose a Dual-Stage Multimodal Alignment approach that integrates breast ultrasound (US) imagery with clinical text reports to improve diagnostic stability. The method proceeds in two stages: (1) Local Correlation Alignment (LCA), which aligns fine-grained visual features with textual embeddings to capture localized lesion attributes, and (2) Global Attention Alignment (GAA), which applies multi-head self-attention to the joint visual–textual sequence to encourage domain-invariant representations. We evaluate the approach on a harmonized, leakage-free repository of 6880 images aggregated from six public US datasets (BUS-CoT, BrEaST, BUS-BRA, BUS-UCLM, BLUI, BUSI) under three protocols: independent benchmarking on BUS-CoT, pooled cross-dataset evaluation, and zero-shot domain generalization on unseen unimodal target domains. On the BUS-CoT benchmark, the 198M-parameter model reaches 0.8177 accuracy and 0.8852 AUC, on par with the 7-billion-parameter Qwen2.5-VL-7B with chain-of-thought reasoning (0.8064 accuracy, 0.8354 AUC) while using roughly 1/35 the parameter count. In the pooled setting, it is competitive with single-domain state-of-the-art methods on individual subsets (e.g., 0.9576 AUC on BUSI, 0.8741 accuracy on BUS-BRA). Under zero-shot transfer without clinical text, per-domain AUC ranges from 0.7360 to 0.8060 across four unseen targets, providing a lower bound under cross-scanner shift. These results indicate that task-specific multimodal alignment can rival large vision-language models in breast US diagnosis at a fraction of the parameter count. Full article

Geovisualizations based on open data are increasingly used as public-facing interfaces for communicating geospatial information, yet their evaluation often remains limited to isolated design, usability, or technical aspects. This study addresses that gap by developing and applying an integrative evaluation framework that combines four analytical dimensions: cartographic representation, interaction and engagement affordances, openness, and accessibility, while treating contextual characteristics as conditioning factors. The framework is operationalized through a mixed-methods content analysis of 26 publicly available geovisualizations based on open data. The results show that most cases are produced by public-sector actors, focus on environmental and transport themes, and rely on conventional cartographic techniques combined with medium levels of interactivity that support structured exploration rather than deeper analytical reasoning. Although many geovisualizations cite data sources and provide some form of data access, licensing remains inconsistent, particularly for the visualization artefacts themselves, limiting reuse potential. Accessibility is implemented even less consistently across geovisualizations, with recurring shortcomings in color contrast, keyboard navigation, screen-reader compatibility, and multilingual support. Overall, the findings suggest that the broader societal potential of geovisualizations based on open data may not be determined by individual features, but by balanced cross-dimensional configurations. Strengthening the integration of openness and accessibility alongside interaction and design may enhance the potential of geovisualizations to support reuse, inclusiveness, and public engagement. Full article

With the rapid development of UAV and satellite remote sensing, in combination with deep learning, high-efficiency monitoring of pine wilt disease (PWD) for forest health management is now feasible. Accurate detection has not yet been realised. The sensing platforms have different ranges of space, observation areas and imaging orientations. At the same time, the target groups for PWD often have weak phenotypic features, are easily affected by a complex forest background, and show irregular data distributions at different stages of the disease. The above factors are limits to the performance of traditional methods based only on general visual features. To address the problems mentioned above, we propose the cross-platform semantic-consistent and phenotype-adaptive detection network SCPA-Net for high-precision PWD detection in both UAV and satellite images. First, we construct a cross-platform multimodal framework to integrate remote sensing images and disease-related text descriptions. The above design adds semantic prior knowledge to expand the model’s capacity for high-level phenotypic attribute extraction without direct observation. Second, to reduce the semantic gap caused by the different platforms, improve the semantic consistency of UAV and satellite images, strengthen discriminative feature channels and salient regions, and address cross-platform misalignment. Third, since the targets are often associated with complex forest environments, target-context relational modeling is enhanced and irrelevant interference is suppressed to reduce the impact of non-causal attributes. As pine wilt disease symptoms gradually progress from mild to severe (e.g., crown discoloration, texture variation, and wilting severity), differences among disease stages may lead to learning imbalance and knowledge forgetting; therefore, a staged adaptation strategy has been proposed. First, the model learns from relatively easy examples. Subsequently, it progressively learns from more difficult examples to enhance generalization performance. Experiments have been conducted on a self-built cross-platform dataset, a satellite dataset, the PDT public dataset, and the Roboflow dataset, and the proposed method has achieved better detection accuracy and generalization. The framework can address the problem of PWD detection in challenging-to-process forestry remote sensing data reasonably well. Full article

Background: Surgical operations are an integral part of healthcare delivery and impose a substantial clinical and operational burden. Understanding how the operation- and process-management literature in healthcare reflects the intensity of surgical services and how this may affect the specialization preferences of healthcare professionals is important for strategic workforce planning. Methods: A bibliometric analysis was conducted on 272 academic publications obtained from the Web of Science Core Collection with the keywords “lean philosophy”, “health” and “process” to capture the operational and process-improvement perspective of healthcare services. In this work, the “lean philosophy” keyword was taken to denote the operation- and process-management view of healthcare services, not to reflect the whole literature on surgical intensity. This selection was performed due to the multiple reasons, with an example being that lean-related studies often discuss complexities of workflow, efficiency, organizational responsiveness, and quality optimization, which are aspects also directly linked to surgical operational intensity. The data were analyzed using the bibliometrix R package, R-4.6.0 to construct the keyword co-occurrence network. Based on this network, a cognitive map was designed to visualize the conceptual relationships among the themes. Thematic clusters based on the co-occurrence network were then evaluated and prioritized by using the Analytic Network Process (ANP). Pairwise comparison data were derived from seven experts (surgeons and healthcare managers), and the model was implemented in Super Decisions with consistency ratios below 0.10. Results: The findings of the co-occurrence analysis are five main thematic clusters with surgical intensity themes including Healthcare Services, Quality, Care, Health and Outcomes. The cognitive map shows that Healthcare Services and Quality have the most central positions and structural hubs in the literature, whereas Outcomes is a dimension of great importance in terms of performance. The ANP results show that Quality (limiting weight ≈ 0.21), General Topics (≈0.14) and Management and Leadership (≈0.13) are the most influential sub-themes with regard to surgical operational intensity and, indirectly, to specialization preferences. Conclusions: The findings reveal that quality management, organizational leadership and larger health policy concerns are closely associated with the intensity of operations of surgical services as depicted in the operation- and process-management literature. Healthcare workers might be inclined to relocate to job positions related to quality improvement and leadership in lieu of places with a high direct clinical burden. Such insights can guide the policies of strategic human resource planning and specialization balancing in healthcare systems. Full article

Knowledge graph link prediction is a fundamental task for improving the completeness and reasoning capability of knowledge graphs. In industrial knowledge graph scenarios, missing relations may limit knowledge completion, relational reasoning, and downstream industrial applications. Fault diagnosis is a representative application scenario, where missing relations among fault phenomena, alarm information, fault locations, and fault causes may further affect fault analysis, maintenance decision-making, and industrial knowledge services. Industrial knowledge graphs usually suffer from sparse local structures, imbalanced relation distributions, explicit entity-type boundaries, and highly confusing candidate entities with similar structural or semantic contexts. These characteristics make it difficult for conventional embedding-based or graph neural network-based methods to achieve reliable candidate ranking by relying only on structural propagation or semantic matching. To address these challenges, this study proposes a type-constrained structural–semantic fusion framework with dynamic relation priors for industrial knowledge graph link prediction, and further investigates its application to fault diagnosis. The proposed framework extends a relation-centered graph neural reasoning backbone by generating dynamic relation priors through query-conditioned relation-level graph propagation over a predefined relation graph, thereby enhancing query-specific structural reasoning. It further introduces a semantic projection module to align textual representations of entities and relations with structural representations at the candidate-ranking stage. In addition, relation-category and hierarchy-aware signals are used to modulate relation representations during propagation, while entity-type constraints are incorporated into final scoring and type-constrained hard negative construction. In this way, structural evidence, textual semantic information, and entity-type validity constraints are jointly used for candidate ranking rather than being treated as isolated signals. Experiments are conducted on two public benchmark datasets, WN18RR and FB15k-237, and two industrial knowledge graph datasets in Chinese and English. The Chinese industrial knowledge graph is constructed from fault diagnosis knowledge and is used as a representative application dataset, while the English industrial knowledge graph is used to further evaluate the adaptability of the proposed framework in a related industrial production scenario. The proposed method achieves MRR scores of 0.599 and 0.446 on WN18RR and FB15k-237, respectively, and obtains MRR scores of 0.8532 and 0.7994 on the Chinese and English industrial knowledge graphs. The results demonstrate that the proposed framework improves both general link prediction performance and industrial-domain adaptability, especially in scenarios involving sparse structures, type-constrained candidate validity, and semantically confusing entities, and shows practical potential for fault diagnosis applications. Full article

Background and Objectives: Patient safety at hospital interfaces is shaped by organisational fragility, ethical obligations, and anticipated legal exposure. Reporting, disclosure, and speaking up have been studied separately, yet the way these pressures converge in ordinary hospital work remains insufficiently described. Materials and Methods: We conducted a qualitative study in a public hospital in Romania using semi-structured episodic interviews and the critical incident technique. Twelve clinicians participated: six nurses and six physicians working in intensive care, emergency medicine, general surgery, paediatrics, oncology day care, anaesthesia, obstetrics, and internal medicine/cardiology. Interviews were audio-recorded, transcribed verbatim in Romanian, anonymised, and analysed with the framework method from a critical realist perspective. A secondary cross-case coding of all 12 episodes was used for descriptive analytic displays. Results: Four mechanisms organised the material. First, local stop rules and cross-checks created temporary stability at fragile interfaces such as high-alert medication, patient identification, specimen labelling, and transfer documentation. Second, consent and confidentiality were repeatedly compressed by urgency, compromised capacity, public space, and family pressure; legitimacy depended on explicit reasoning rather than documentary completion alone. Third, speaking-up and near-miss reporting were governed by protocol-backed legitimacy, leader response, and the informal cost of interruption. Formal incident reporting was present in one episode, partial in one, and absent in 10. Fourth, documentation and disclosure redistributed accountability. Notes that recorded reasoning supported continuity of care, whereas protective opacity concealed near misses, infrastructural weakness, and interactional pressure. Documentation or disclosure pressure appeared in all 12 episodes. Conclusions: Safety in everyday hospital work was assembled through local barriers, moral triage, and selective visibility. Interface redesign, protected near-miss reporting, psychologically safe escalation, and structured support for urgent consent and post-incident communication would make transparent safety work more sustainable. Trustworthiness was strengthened through reflexive memoing by the physician-interviewer, an audit trail of coding decisions, comparison across professional groups, active attention to negative cases, and iterative assessment of meaning saturation at the level of explanatory mechanisms. Full article

Smart cities generate vast, heterogeneous data streams from transportation networks, energy grids, environmental sensors, and public services, yet the semantic fragmentation of these data silos prevents urban operators from deriving actionable, cross-domain intelligence. Knowledge graphs (KGs) have emerged as a powerful paradigm for integrating diverse, large-scale data collections through graph-based representations of entities and their relationships. This paper applies the Design Science Research Methodology (DSRM) to design, develop, and evaluate UrbanKG, a layered artifact that deploys knowledge graphs as the semantic backbone of smart city data infrastructure. We demonstrate the framework through a proof-of-concept implementation using publicly available urban datasets across five domains, yielding a 287,000-triple knowledge graph validated through cross-domain SPARQL queries and accessibility analysis. Following the six DSRM process steps—problem identification, objective definition, design and development, demonstration, evaluation, and communication—the framework addresses ontology design, multi-source data fusion, federated governance, temporal reasoning, and hybrid deductive–inductive inference. The artifact satisfies all five design objectives and contributes four transferable design principles. Six open research challenges are identified as the forward research agenda. Full article