Krirk University

The emergence of large-scale models and machine learning has transformed the modeling of complex nonlinear systems, such as postseismic stress evolution. However, purely data-driven approaches often lack interpretability and numerical stability, leading to physically inconsistent long-term predictions. This study addresses these limitations by introducing a coupled Kolmogorov–Petrovsky–Piskunov–Rate-and-State (KPP–RS) reaction–diffusion system as a rigorous physical prior for large-scale modeling of stress-driven dynamics. Using analytic semigroup theory and Banach’s fixed-point theorem, we establish the global existence and uniqueness of solutions, ensuring that the governing dynamics are mathematically well posed—a necessary prerequisite for stable learning-based frameworks. We further prove the global dissipativity of the system and identify a bounded absorbing set in the H1 phase space, which imposes intrinsic physical constraints and limits unphysical parameter exploration in large-scale optimization or black-box modeling. In addition, a Courant–Friedrichs–Lewy (CFL) stability condition is derived, providing a theoretical benchmark for time-step selection in numerical implementations, including physics-informed or hybrid neural architectures. This analytical framework supplies a mathematically controlled foundation for developing robust, interpretable, and stable pattern-recognition or time-series representations in complex geophysical systems.

In buildings, steel-concrete composite beams with web openings save service space but weaken structural performance. To address this issue, this study proposes the use of a perfobond strip (PBL) connector in the opening area to compensate for the loss of connection performance and composite action caused by the web openings. A total of 25 specimens were subjected to experimental testing and finite element simulation to investigate the effects of 11 varying parameters-including connector type, hole radius of the perforated plate, hole position, and the diameter of through reinforcement-on the mechanical performance and composite behavior of the specimens. The results indicate that, compared to traditional stud connectors, PBL connectors increase the shear force transferred into the concrete by 13.57%-18.33%, thereby enhancing the structure's shear resistance. Specimens with PBL connectors in the opening area exhibited favorable ductile failure characteristics and improved shear capacity, with failure modes transitioning to concrete flange splitting and rebar shearing. Among the parameters studied, increasing the hole radius of the PBL plate resulted in the greatest load-bearing capacity, while increasing the diameter of the penetrating rebars most effectively reduced the relative slip behavior of the specimens.

With the recent growth in coastal development, dredged clay has been increasingly used as a core reclamation material. However, marine clay may exhibit a consolidation behavior distinct from that of terrestrial clay owing to pore water salinity. In particular, salinity variations induced by environmental factors such as groundwater level fluctuations or seawater leaching can alter the compressibility and settlement behavior of clay deposits, thereby increasing uncertainty in settlement prediction. In this study, disturbed dredged clay was collected and remolded at different salinity levels (0%, 1%, 2%, and 3%) to investigate the effects of salinity variation on consistency and consolidation responses. The experimental results indicated that decreasing salinity increases the consistency indices (liquid limit and plastic limit) and clay compressibility. These trends are attributed to the expansion of the diffuse double layer and the consequent loosening of the clay microstructure under low-salinity conditions. These findings highlight the importance of considering pore water salinity for reliable settlement prediction and the rational design of dredged coastal reclamation grounds.

Green transition is a must for organizational sustainable development, with employee competence being a key driver. Aiming at the problem of insufficient systematic evaluation of employee competence in existing green transformation research, this paper presents a four - dimension evaluation system (knowledge, skills, attitude, and ability) with ten sub – criteria. To evaluate employee competence in green transition enterprises, this paper proposes a competence evaluation method, based on probabilistic hesitant fuzzy sets (PHFS). The proposed method integrates SWARA and the exponential probabilistic hesitant fuzzy entropy for criteria weight quantification, balancing expert experience and data - driven approaches. It handles the uncertainty of evaluation through probabilistic hesitant fuzzy sets, and uses the CoCoSo method for comprehensive ranking, which enhances the scientific and operability of evaluation. To test the proposed method, it is applied to evaluate employee competence in a green manufacturing company and compared with Shapely - TODIM. The consistent ranking results from both methods confirm the scientific and practical applicability of the proposed method.

ABSTRACT Due to climate change and government economic pressures, understanding Public Green Finance (PGF) factors is vital. Despite worldwide sustainability promises, less is known about how technological, economic, and policy factors drive public green investments, especially in industrialized economies. We fill that gap by empirically evaluating the primary determinants of PGF in the G7 countries, where innovative capacity and environmental policy frameworks drive low‐carbon growth. The article under investigation examines the determinants of PGF over the period 1995–2022 in G7 countries that refer to the roles of the important variables: Environmental Technology (ET), Energy Efficiency (EE), Environmental Policy Stringency (EPS), Economic Development (ED), Trade (TR), Domestic Energy‐Related Uncertainty (DEUI), Human Capital (HC), Domestic Economic Policy Uncertainty (DEPU), and Public Green Finance (PGF). To adopt the best econometric methods, such as Panel Two‐Stage least Squares (2SLS) and Driscoll–Kraay Fixed Effects regression, this study involves the endogeneity, autocorrelation, and heteroskedasticity to derive strong and dependable findings. The results show that higher EPS, ED, and TR have a large positive impact on PGF, whereas higher EE and DEUI have negative links with PGF. Furthermore, HC is pointed out as being a significant element that can be used in improving the efficacy of the PGF initiatives. The paper also shows the subtle effects of DEPU that may either lead to encouragement or discouragement of green investments based on the setting. The outcomes of these studies can be useful in informing policymakers, financial intermediaries, and stakeholders in G7 nations, which is part of a larger agenda of promoting sustainable development and mitigating the issue of climate change. This study also corresponds with the Sustainable Development Goals (SDG) established by the United Nations, especially SDG 7, SDG 9, and SDG 13, which provide strategic value to positively mobilize green finance in the G7 countries.

Online learning environments tend not to provide the social and pedagogical cues of physical classrooms, so evaluating student engagement and emotional states in real time becomes challenging. Current methods depend mainly upon facial expression recognition or textual sentiment analysis, constraining the depth and accuracy of behavioral interpretation. This research suggests a multimodal learning analytics framework that combines visual and textual data to infer learner emotions and engagement for improving the interpretability, responsiveness, and pedagogical value of learning analytics systems in digital education. Two datasets were created: (a) a facial expression dataset of 10,000 grayscale images annotated over five emotion categories and (b) an engagement dataset of 4,000 images annotated according to behavioral indicators. Concurrently, 1,667 learner feedback responses from massive open online courses were prepared for sentiment analysis. Convolutional neural networks (CNNs) were used for emotion and engagement classification, and a fine-tuned BERT (bidirectional encoder representations from transformers) model for sentiment analysis. A rule-based integration engine combined outputs to create multidimensional behavioural typologies. The CNN models reached >92% validation accuracy for both emotion detection and engagement detection tasks, whereas the BERT sentiment classifier achieved F1 = 0.87 and 88.1% accuracy. The multimodal integration procedure identified four unique learner behavior typologies (e.g., students who were cognitively engaged but visually disengaged). The framework offers an accurate, interpretable, and scalable real-time learning analytics solution. Compared with previous methods, it overcomes significant limitations and offers a useful resource for facilitating adaptive, data-based instruction interventions, especially in online and health science education.

With the rapid expansion of global Chinese language education, traditional instructional approaches have become insufficient to address the needs of diverse and cross-cultural learners. Systematic and learner-centered instructional design has therefore become an increasingly important topic in international Chinese language education. This study develops and validates an integrated model that incorporates the ADDIE instructional design framework, self-efficacy, and creative climate, aiming to examine how systematic instructional design influences international Chinese learners’ learning outcomes and learning satisfaction, as well as the psychological mediating mechanism and contextual moderating effect underlying this process. Based on 466 valid questionnaire responses, structural equation modeling was employed for analysis. The results indicate that: (1) the implementation of the ADDIE model significantly enhances learning outcomes and learning satisfaction, demonstrating the effectiveness of systematic instructional design in international Chinese teaching; (2) self-efficacy plays a significant mediating role between ADDIE and learning outcomes, suggesting that systematic instructional design can improve learning performance by strengthening learners’ confidence and sense of capability; and (3) creative climate positively moderates the relationship between self-efficacy and learning outcomes, indicating that an open, supportive, and exploratory learning environment amplifies the positive influence of self-efficacy. Theoretically, this study extends the applicability of the ADDIE model to the field of language education and reveals the interaction among instructional design, learning psychology, and learning context. Practically, the findings provide strategic implications for curriculum development, learner support, and classroom innovation in international Chinese language teaching. Finally, directions for future research are proposed to further deepen the understanding of systematic instructional design in cross-cultural learning settings.

Chronic and acute pancreatitis (CP and AP, respectively) are debilitating conditions with significant morbidity and mortality, necessitating a comprehensive understanding of their underlying mechanisms. This study provides a high-resolution, multi-omics investigation into the genetic and immune cell underpinnings of pancreatitis, integrating rare familial CP with a large cohort of patients with AP. Utilizing an integrative approach that combined whole-exome sequencing (WES) from two pediatric CP patients and their family members with single-cell RNA sequencing (scRNA-seq) and bulk transcriptomics from a public AP cohort (n = 119), we identified a shared molecular and cellular pathology. WES of the CP family revealed heterozygous mutations in 12 novel genes, including EXOC4, ATG2A, and UNC80. Functional enrichment analysis highlighted autophagy, cell adhesion, and vesicle-mediated transport as the key biological processes implicated in the pathophysiology of both conditions. Single-cell profiling of peripheral blood mononuclear cells (PBMCs) from the CP family revealed a marked increase in the proportion of naive B cells and an altered activity of CD8+ T cells, suggesting a dysregulated B-cell-mediated immune response. This observation was corroborated in the AP cohort, where CIBERSORT analysis revealed a significant increase in both naive B cells and CD8+ T cells correlating with the disease severity. Weighted gene co-expression network analysis (WGCNA) on the AP cohort uncovered 14 gene modules associated with disease progression. These modules were significantly enriched for pathways central to the innate immune response, including complement-dependent cytotoxicity and neutrophil degranulation, providing a molecular link to the observed immune cell infiltration. An artificial intelligence (AI)-driven model incorporating 110 CP family-related genes (GTCPFs) demonstrated exceptional predictive capability (average AUC > 0.84) for AP severity, highlighting the translational potential of our findings. The model identified a robust signature of 17 genes, including ATG2A, EXOC4, and TNS1, which may serve as novel diagnostic and prognostic biomarkers. Our findings provide a unified view of the pathogenesis of pancreatitis, linking novel genetic variants to specific immune cell and transcriptomic signatures. This integrative approach underscores the critical importance of both genetic and immune factors in CP and AP, identifying potential biomarkers and therapeutic targets and paving the way for personalized medicine in the management of these challenging conditions.

Criminal networks feature complex structures and high internal concealment. Social computing methods and deep learning provide a possibility to identify hidden key figures and predict criminal behaviors within them. To this end, a Key Person Identification Model Based on Account Association (KPI-AA) is proposed. This model combines local neighbor similarity and global edge betweenness, and uses graph neural network (GNN) to conduct in-depth characterization of key nodes in social networks. It is applied to reveal potential organizational cores in complex social structures and identify potential diffusion paths in criminal networks. Experimental results show the following: In the dimension of propagation dynamics, KPI-AA infects 34 nodes at 40-time steps in the Zachary network, and reaches 329.6 nodes at 150-time steps in the Harry Potter network. Both results are higher than those of baseline models. In terms of network robustness, after removing 70% of the nodes in the Zachary network, the relative connectivity of KPI-AA is only 0.087. Regarding ranking consistency, the Kendall's tau coefficient on the Twitter dataset reaches 0.467. Computational efficiency analysis indicates that while maintaining performance advantages in propagation dynamics, network robustness, and ranking consistency, KPI-AA still exhibits excellent scalability and practical deployment feasibility. The above results indicate that KPI-AA has advantages in propagation speed, revealing network vulnerability, and ranking consistency. Therefore, the KPI-AA model is practical in identifying core members of criminal networks and predicting criminal behaviors, and can serve social security governance and criminal investigation applications.

Students diagnosed with benign tumors often experience psychological stress and social isolation, which can adversely affect their emotional well-being. Traditional sports games, which are culturally important and promote physical activity, offer a potential intervention to alleviate stress and improve social support. The research discovers the role of traditional sports games in relieving psychological stress and improving social support for students with benign tumors. A total of 131 students diagnosed with benign tumors were divided into two groups: the intervention group (IG) (n = 91), which participated in traditional sports games, and the Control group (CG) (n = 40), which continued with their regular activities without engaging in sports games. The Perceived Stress Scale (PSS) was utilized to evaluate psychological stress, while the Social Support Scale (SSS) measured the level of social support received by participants, concentrating on emotional, informational, and instrumental support. The statistical analysis utilized IBM SPSS software version 29. The data were analyzed utilizing statistical methods, including descriptive statistics, paired t-tests, and correlation analysis. The IG attained positive outcomes across all factors, with emotional support at 90%, the group perceived cohesiveness at 88%, self-efficacy at 80%, 82% for informational support, 75% for instrumental support, and 68% for psychological stress1. Traditional sports games and team bonds facilitate positive interactions among students. This mode promotes emotional interaction, facilitating social interaction where players support and encourage each other through play, which helps them communicate emotionally. Consequently, these enjoyable social dynamics facilitate lowering psychological stress and a sense of community among students having benign tumors, which enhances our students’ psychological well-being. The results indicate that the IG displayed an important reduction in psychological stress, with developed PSS and SSS scores. Reductions in stress and anxiety, as well as increases in self-efficacy, were significantly correlated with group perceived cohesiveness. Traditional sports games help reduce stress and improve social support for students with benign tumors by fostering peer relationships, building a sense of community, and improving emotional well-being with health-focused programs.