This study investigated the effectiveness of Daniel's Model in teaching cybersecurity concepts to ninth-grade students in Amman, Jordan. A quasi-experimental design with two groups was employed; the experimental group received instruction based on Daniel's model, while the control group followed traditional teaching methods. The sample consisted of 120 students, was selected from ten international schools via convenience sampling, and was randomly assigned to the groups. Data were collected using a 33-item knowledge test covering four key domains of cybersecurity. Analysis in SPSS included an independent samples t-test and a paired samples t-test; in addition, effect size (η²) was calculated to understand the magnitude of the model impact. A higher score of digital threat knowledge and skills in terms of prevention, ethical behaviors, and data protection related to the model taught was noticeable among the experimental group, since the effect size (η²) varied from large (0.34) to very large (0.62). Also, in the delayed test, the experimental group showed no decline in performance, demonstrating that the model was effective in teaching knowledge and digital practices. Considering the results, it is recommended that Daniel's Model, combined with interactive techniques and scenario building, be integrated in teaching digital ethics and use of technology, in addition to training teachers on systematic implementation of desired, sustained outcomes of the model.

This study aims to examine the causal impact of macroeconomic uncertainty on banking stability across six ASEAN emerging markets from 2010 to 2023, with particular attention to structural regime shifts triggered by the COVID-19 pandemic. To achieve this objective, a novel country-specific uncertainty index is constructed using Principal Component Analysis (PCA) based on three indicators World Uncertainty Index (WUI), World Pandemic Uncertainty Index (WPUI), and World Sentiment Index (WSI). Employing advanced causal inference methods, including Double Machine Learning (DML) and Causal Forests, the study estimates both Average Treatment Effects (ATEs) and Conditional Average Treatment Effects (CATEs). The results reveal that a one-unit rise in macroeconomic uncertainty reduces the Z-Score by 10.7% on average, signaling increased financial instability. The adverse effect is most pronounced for small banks (21.9% decline), reflecting limited capital buffers and structural vulnerability, and becomes more severe after the COVID-19 outbreak. CATEs results highlight significant cross-country heterogeneity, with Singapore and Thailand showing resilience, while Indonesia and the Philippines exhibit greater fragility. This study contributes to the literature by integrating SHAP-based model interpretability into causal machine learning for banking stability analysis, offering novel, policy-relevant insights for uncertainty management in emerging ASEAN economies.

This study compares machine learning and econometric approaches for forecasting agricultural export values in volatile global markets, examining predictive accuracy and economic interpretability trade-offs. Monthly data from January 2014 to December 2023 were analyzed using five models: Artificial Neural Networks (ANN), Long Short-Term Memory (LSTM), Hybrid ANN-LSTM, Ordinary Least Squares (OLS), and Autoregressive Distributed Lag (ARDL). Key predictors included durian, mangosteen, and longan export values/volumes, plus China's GDP. Performance evaluation used MAE, RMSE, MAPE, and R² metrics with systematic hyperparameter optimization through grid search and 5-fold cross-validation. ANN achieved the highest absolute accuracy (MAE: 1,684,667,401.55; RMSE: 2,602,671,952.28), while Hybrid ANN-LSTM delivered superior relative accuracy (MAPE: 1.58%). ARDL demonstrated exceptional explanatory power (R²=0.83) for structural economic relationships. China's GDP emerged as the strongest determinant across all models. Longan export value showed contrasting effects between approaches, positive in machine learning models versus negative in econometric models, reflecting different paradigmatic interpretations of market substitution dynamics. This research introduces the first comprehensive comparative framework integrating advanced hybrid neural networks with traditional econometric methods for multi-commodity agricultural forecasting, addressing cross-commodity substitution effects previously unexplored while offering complementary perspectives for both predictive accuracy and economic policy interpretation.

Abstract This study investigates the determinants of capital structure by comparing firms listed on two prominent global stock indices: the S&P 500 (United States) and the NSE CNX 500 (India). Specifically, it examines how firm-specific factors—such as liquidity, asset tangibility, and sustainability practices—influence leverage decisions within differing economic and institutional contexts. Drawing on a comprehensive dataset of 3,575 firm-year observations from 406 S&P 500 companies and 4,180 observations from 419 NSE CNX 500 firms between 2011 and 2021, the analysis employs Two-Stage Least Squares (2SLS) regression, the Generalized Method of Moments (GMM), and a series of diagnostic tests addressing heteroskedasticity and model robustness. The empirical results indicate that liquidity, tangibility, and sustainability performance significantly affect firms’ capital structure decisions. Moreover, growth opportunities and profitability also play key roles. Cross-country differences highlight the influence of macroeconomic conditions and financial system structures on leverage behavior. This research enriches the capital structure literature by offering a comparative, cross-national perspective and provides actionable insights for corporate managers, investors, and policymakers seeking to optimize capital structure in diverse financial environments

This study examines how technology-organization-environment (TOE) readiness (TR), ambidextrous organization (AO), and organizational resilience (OR) influence innovation performance (IP) and competitive advantage (CA) in small and medium-sized enterprises (SMEs), focusing on craft-based firms operating in dynamic and uncertain environments. A quantitative research design was chosen to collect survey data from 200 Indonesian craft-based SME owners and managers. Structural equation modelling (SEM) was employed to analyze the data and examine both direct and indirect relationships among the proposed constructs. The findings demonstrate that TR, AO, and OR each have a significant positive effect on IP. In turn, IP has a strong positive effect on CA, highlighting its key role in converting organizational capabilities into market success as well as a complementary mediating mechanism linking TR, AO, and OR to competitive outcomes. The novel contribution of this study is the empirical integration of TR with internal dynamic capabilities, namely AO and OR, in a single structural model that explains how SMEs transform innovation-related capabilities into CA. This integrative perspective contributes to the literature on SME innovation and offers practical implications for managers and policymakers seeking to strengthen innovation-driven competitiveness and long-term sustainability amid turbulent environmental conditions in emerging market contexts.

This study developed a novel hybrid Graph Convolutional Network–Long Short-Term Memory (GCN–LSTM) model to forecast greenhouse gas (GHG) emissions across multiple country sectors, aiming to enhance climate policy. We analyzed 52 years (1970–2022) of GHG emissions data (CO₂, CH₄, N₂O, F-Gases) from 163 countries and eight sectors (Agriculture, Buildings, Fuel Exploitation, Industrial Combustion, Power Industry, Processes, Transport, Waste), sourced from the EDGAR v8 database. The GCN adjacency matrix captures spatial relationships on a weighted sum of Haversine distance and cosine similarity, while the LSTM models temporal dynamics. Data preprocessing includes min-max scaling and outlier handling with Interquartile Range capping. The model was trained on 70% of the data, validated on 15%, and tested on 15%, using Mean Squared Error (MSE) loss and the Adam optimizer. The performance was evaluated with Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Coefficient of Determination (R²). The GCN–LSTM model outperformed baseline models (ARIMA, Simple LSTM, Stacked LSTM), achieving the lowest MAE (0.0207 in Waste) and highest R² (0.9756 in Waste). Model interpretability highlighted strong regional connections, such as Thailand–Cambodia in the Waste sector, suggesting that spatial and temporal dependencies offer superior forecasting accuracy, informing targeted climate action.

Objectives: This study designs and evaluates a social intelligence-based learning model for social studies in modern Islamic boarding schools (pesantren) in Indonesia, addressing the lack of social-emotional integration in faith-based curricula. Methods/Analysis: A mixed-methods approach involved 150 participants from five pesantren, including students, teachers, and school leaders. Qualitative data were gathered via interviews, observations, documentation, and FGDs, while quantitative data were collected using a validated 25-item Social Skills Questionnaire. Descriptive statistics, paired-sample t-tests, and Cohen’s d assessed effectiveness, with methodological and theoretical triangulation ensuring qualitative validity. Findings: The model—integrating social values, interactive strategies, contextual materials, authentic assessment, and teacher facilitation produced significant gains in social intelligence. Problem-solving and tolerance improved most (+1.7 points each), and student participation rose from 42% to 88%. Both students and teachers reported high satisfaction with the collaborative, contextually relevant approach. The questionnaire demonstrated high reliability (Cronbach’s α = 0.87). Novelty/Improvement: This study introduces a culturally grounded, faith-based pedagogical framework embedding social intelligence into character-based education. By aligning Islamic values and Acehnese wisdom with 21st-century social competencies it provides a replicable model for enhancing social-emotional learning in similar educational contexts.

This study aims to develop an explainable machine learning–based early warning system to predict dropout risk among Science, Technology, Engineering, and Mathematics (STEM) students in the MENA region. Using longitudinal data from 6,798 undergraduate STEM students enrolled at a major UAE university, we evaluated six supervised classifiers: XGBoost, Gradient Boosting Machine (GBM), Random Forest, CART, Logistic Regression, and K-Nearest Neighbors. Models were trained on institutional student information system (SIS) data spanning ten cohorts (2010–2019), with class imbalance addressed through ROSE sampling. The top-performing models (XGBoost, GBM, and Random Forest) achieved AUC-ROC scores exceeding 0.91 and F1-scores above 0.84, significantly outperforming baseline models. Key predictors of dropout included the number of withdrawn semesters, second-term credit load, academic probation history, and performance in mathematics and physics. To improve interpretability, we applied SHapley Additive exPlanations (SHAP) analysis, enabling both global and individual-level feature attribution. The system offers scalable, real-time predictive capabilities using only routinely available SIS data, with no need for external surveys or learning management system inputs. The novelty of this research lies in its integration of explainable AI into a regional context, enabling early, transparent, and actionable interventions to reduce dropout. These findings contribute to data-driven retention strategies in higher education systems where predictive tools remain underutilized.

This study examines the role of credit allocation to the private sector in driving economic growth in Vietnam’s transitional economy. The primary objective is to evaluate whether bank credit allocation fosters sustainable output expansion or, conversely, produces diminishing returns when it surpasses optimal levels. Employing the Autoregressive Distributed Lag (ARDL) bounds testing framework, the analysis uses annual data for 1990–2024 and compares three specifications: credit to the private sector, aggregate credit to the economy, and credit to the state sector. Findings indicate a robust long-run cointegration between credit and output, but with a clear nonlinear pattern: private credit enhances growth up to a threshold of roughly 91% of GDP, beyond which its marginal effect declines. While capital formation and moderate inflation consistently support long-term growth, foreign direct investment exerts mainly short-term benefits, and state-directed credit shows no significant contribution. The novelty of this paper lies in extending previous studies through a longer time horizon, updated post-GDP-revision data, and explicit disaggregation between private and state credit. By highlighting threshold effects and sectoral inefficiencies, this research improves understanding of the credit–growth nexus in transitional economies and underscores the need to prioritize credit quality, efficiency, and SME access in credit policy. JEL Classification: E51, G21, O47, P27.

Large Language Models (LLMs) have shown impressive capabilities in natural language understanding and generation tasks. However, their reliance on text-only input limits their ability to handle tasks that require multimodal reasoning. To overcome this, Multimodal Large Language Models (MLLMs) have been introduced, enabling inputs such as images, text, video and audio. While MLLMs address some limitations, they often suffer from hallucinations because of over-reliance on internal knowledge and face high computational costs. Traditional vector-based multimodal RAG systems attempt to mitigate these issues by retrieving supporting information, but often suffer from cross-modal misalignment, where independently retrieved text and image content cannot align meaningfully. Motivated by the structured retrieval capabilities of text-based knowledge graph RAG, this paper proposes VisGraphRAG to address the challenge by modelling structured relationships between images and text within a unified MMKG. This structure enables more accurate retrieval and better alignment across modalities, resulting in more relevant and complete responses. The experimental results show that VisGraphRAG significantly outperforms the vector database-based baseline RAG, achieving a higher answer accuracy of 0.7629 compared to 0.6743. Besides accuracy, VisGraphRAG also shows superior performance in key RAGAS metrics such as multimodal relevance (0.8802 vs 0.7912), showing its stronger ability to retrieve relevance information across modalities. These results underscore the effectiveness of the proposed Multimodal Knowledge Graph (MMKG) methods in enhancing cross-modal alignment and supporting more accurate, context-aware generation in complex multimodal tasks.