patterns of behavior they personally consider to be violent. It is precisely this subjective perception of named and reported, or they are normalized as part of a teacher's job, even though, in the long run, they can negatively affect a teacher's professional identity and well-being. A phenomenographic perspective was applied in this research. It has been found that teachers understand violence in the workplace as are characterized by verbal, physical, written, and visual expressions. The scope of the research results shows that the teachers' perception of violence is not limited to incidents but is revealed as a process of that the teachers' understanding of violence is more related to administrative power practices.
Accurate and efficient land-use and land-cover (LULC) classification from remote sensing imagery remains challenging. This is because it requires capturing long-range spatial dependencies while maintaining computational scalability. Recent transformer-based models improve global context modeling. However, they suffer from quadratic complexity and are limited in applicability to high-resolution imagery. We introduce Mamba-RSI: a linear-time, state-space deep learning framework using selective recursion, hierarchical multi-scale feature extraction, and lightweight global representations. Mamba-RSI captures both fine-grained spectral/texture information and coarse structural patterns with significantly less computational overhead than existing quadratic self-attention transformers. Extensive experimentation on EuroSAT and NWPU-RESISC45 demonstrated that Mamba-RSI achieves state-of-the-art performance. It achieved 99.72% accuracy on EuroSAT and 96.84% on RESISC45. This represents a +0.40% improvement over the strongest transformer baseline, ATMformer, on EuroSAT, a +0.29% improvement on RESISC45, and more than +0.53% over ViT-B on EuroSAT. Robustness tests under severe Gaussian noise (sigma = 0.10) showed that Mamba-RSI maintains 97.43% accuracy. MaxViT, by comparison, maintains 94.01% in the same setting. Mamba-RSI also preserves 91.15% accuracy under 30% patch occlusion, outperforming ViT-B by +7.41%. Mamba-RSI provides an attractive blend of accuracy, robustness, and efficiency. It serves as a scalable foundation for new insights into remote sensing analytics and LULC mapping systems.
Land degradation is a global ecological and environmental issue to which the drylands with vulnerable socioecological systems are particularly sensitive. Ecological restoration is regarded as a vital approach to halt, curb or reverse land degradation. However, the lack of accurate information on land degradation dynamics poses a significant barrier to developing effective restoration strategies. Here, we selected fourteen typical land degradation processes in China's drylands, employed a generalised additive model to identify degradation thresholds and classify degradation states, and analysed degradation trends using the Theil-Sen slope statistic and Mann-Kendall test. We then applied the convergence of evidence method to conduct a comprehensive, multidimensional assessment coupling degradation states and trends, ultimately identifying priority areas for ecological restoration. The results showed that aridification, loss of soil organic carbon, and soil alkalisation are the most widespread forms of land degradation facing China's drylands. 70.33 % of the area is affected by 1 to 7 types of land degradation (mildly degraded state), 25.63 % is affected by 8 to 14 types (severely degraded state), and 4.04 % shows no signs of degradation (non-degraded state). While land degradation across all three states generally exhibits a stable trend, mildly degraded areas showed signs of improvement, whereas severely degraded areas remain at risk of further degradation. Based on the states and trends of land degradation, China's drylands can be categorised into restoration-priority areas (6.79 %), conservation-priority areas (11.74 %), and management-priority areas (81.47 %), each with distinct focus and intervention needs. This study can help decision-makers understand the dynamics of land degradation and develop priority strategies for ecological restoration.
This article examines the ideological repurposing of institutions of higher management education (IHMEs) from a systems-theoretical perspective. Drawing on Niklas Luhmann's theory of social systems, we analyse both historical and contemporary cases to identify structural similarities in how institutions of higher education (IHEs) have been reoriented to serve external political imperatives. Through a functional comparison of Soviet, Nazi and nationalist regimes with current trends in sustainability- and DEI-driven transformations, we argue that repurposing efforts operate through similar mechanisms: modifications of personnel structures, communication channels, decision programmes and organisational culture. We conceptualise IHE as multifunctional organisations that mediate structural couplings between various function systems, particularly science, education and politics. Our analysis shows that when decision programmes become aligned too tightly with the logic of powerful political organisations, IHEs risk losing their operational autonomy and functional distinctiveness. The paper concludes by warning that even well-intentioned missions may lead to epistemic closure and ideological totalisation if historical lessons are neglected.
Background/Objectives: Accurate and reliable automated dermoscopic lesion classification remains challenging. This is due to pronounced dataset bias, limited expert-annotated data, and poor cross-dataset generalization of conventional supervised deep learning models. In clinical dermatology, these limitations restrict the deployment of data-driven diagnostic systems across diverse acquisition settings and patient populations. Methods: Motivated by these challenges, this study proposes a transformer-based, dermatology-specific foundation model. The model learns transferable visual representations from large collections of unlabeled dermoscopic images via self-supervised pretraining. It integrates large-scale dermatology-oriented self-supervised learning with a hierarchical vision transformer backbone. This enables effective capture of both fine-grained lesion textures and global morphological patterns. The evaluation is conducted across three publicly available dermoscopic datasets: ISIC 2018, HAM10000, and PH2. The study assesses in-dataset, cross-dataset, limited-label, ablation, and computational-efficiency settings. Results: The proposed approach achieves in-dataset classification accuracies of 94.87%, 97.32%, and 98.17% on ISIC 2018, HAM10000, and PH2, respectively. It outperforms strong transformer and hybrid baselines. Cross-dataset transfer experiments show consistent performance gains of 3.5-5.8% over supervised counterparts. This indicates improved robustness to domain shift. Furthermore, when fine-tuned with only 10% of the labeled training data, the model achieves performance comparable to fully supervised baselines. Conclusions: This highlights strong data efficiency. These results demonstrate that dermatology-specific foundation learning offers a principled and practical solution for robust dermoscopic lesion classification under realistic clinical constraints.
