ABSTRACT. This paper presents a detailed investigation into the fundamental principles governing fluid dynamics, with a specific focus on the role of volumetric forces, conservation laws, and viscous stresses. Through a comprehensive theoretical framework, I explore how these forces and stresses influence the behavior of fluids in various conditions, from laminar to turbulent flow regimes. The conservation laws, including mass, momentum, and energy, are systematically derived and applied to both Newtonian and non-Newtonian fluids, providing insights into the complexities of fluid motion. Additionally, the paper examines the contribution of viscous stresses to the internal frictional forces within fluids, highlighting their significance in the development of flow resistance and boundary layer formation. Computational simulations are employed to validate the theoretical models, offering a robust analysis of the fluid dynamics in different scenarios. The results underscore the critical importance of accurately accounting for these factors in fluid mechanics, with implications for both fundamental research and practical applications in engineering disciplines.

ABSTRACT. The rapid growth of social commerce platforms such as Shopee, Lazada, and TikTok Shop has transformed consumer decision-making, yet the psychological mechanisms driving sustainable repurchase remain underexplored. Previous studies show that the Fear of Missing Out (FOMO) encourages impulsive purchases (Przybylski et al., 2013; Good & Hyman, 2020), while the Fear of Exclusion (FOE) fosters conformity and compensatory consumption (Zhang, 2025; Jiang, Lee, Jin, & Kan, 2025). However, little is known about how these constructs interact to influence long-term loyalty. Drawing on the Stimulus–Organism–Response paradigm (Mehrabian & Russell, 1974) and the Stressor–Strain–Outcome framework (Jeong, 2022), this study develops and tests a model in which personal traits, influencer credibility, and positive information (Lou & Yuan, 2019; Masuda, 2022; Chen, Lu, Wang, & Pan, 2019) drive FOE and FOMO, which subsequently affect repurchase and loyalty. Social commerce engagement (SCE) is conceptualized as a moderator that strengthens these psychological effects. Survey data from 385 Vietnamese consumers were analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM). Results show that FOE amplifies FOMO and directly enhances repurchase, while FOMO strongly predicts both repurchase and loyalty. Repurchase, in turn, emerges as a proximal outcome leading to sustainable loyalty (Oliver, 1999; Molinillo, Anaya-Sánchez, & Liébana-Cabanillas, 2020). This study contributes by integrating FOE and FOMO into a unified framework and by identifying SCE as a novel boundary condition, advancing both SOR and SSO theories. Unlike prior research examining these constructs separately, this study highlights their combined effects, offering new insights for theory and practical strategies for platforms to design responsible FOMO-based campaigns, foster engagement, and build enduring loyalty.

ABSTRACT. In an age when artificial intelligence (AI) and digital technologies shape how individuals learn, communicate, and make decisions, education faces an urgent ethical challenge: preparing learners to use technology wisely, responsibly, and compassionately. This dissertation investigates how educational institutions can cultivate ethical digital citizenship—a form of digital participation grounded in awareness, responsibility, and moral reasoning. The research explores the intersection of AI literacy, digital ethics, and moral education, focusing on how schools and universities can embed ethical understanding within digital learning environments. Using a mixed-methods approach, the study combines survey data from 100 students and interviews with 10 educators across Vietnamese institutions. The findings reveal a significant gap between students’ digital competence and their ethical awareness, especially concerning issues such as algorithmic bias, plagiarism in AI-generated content, and data privacy. Drawing from these insights, the study proposes a three-pillar framework for Ethical Digital Citizenship Education (EDCE)—comprising Ethical Awareness, Responsible Participation, and Reflective Judgment. Implementing this framework can support educators in designing curricula that integrate AI ethics with critical thinking and civic responsibility. Ultimately, this dissertation argues that building an ethical digital future begins in the classroom. Education must not only teach how to use technology but also how to question it, ensuring that the next generation

ABSTRACT. In a world increasingly defined by overconsumption and digital immersion, artificial intelligence companions, such as Replika and ChatGPT’s friend mode, are transforming how individuals forge connections, make purchasing decisions, and embrace sustainable practices. This study investigates the potential of these virtual relationships to reduce impulsive buying, enhance mental well-being, and promote eco-friendly lifestyles across different generations, including Generation Z, Millennials, and older adults. Using a mixed-methods approach, we surveyed 407 participants and conducted 8 in-depth interviews, guided by theories of Self-Determination, Compensatory Consumption, and Technology Acceptance Model. The research links artificial intelligence, sustainability, and emotional health, offering practical insights for designing ethical AI tools and crafting policies that encourage meaningful and balanced digital lives.

ABSTRACT. Vietnam is one of the world’s largest coffee exporters, yet smallholder farmers in Vietnam’s Central Highlands remain highly exposed to farm-gate price volatility and information asymmetries in transactions with traders. This paper develops a hybrid forecasting and fairness‐oriented framework to support dynamic and more equitable coffee pricing in Vietnam. Using 108 monthly observations for Đắk Lắk Province from 2016–2024, we assemble a panel of farm-gate coffee prices together with export volume and value, consumer prices, climate variables (rainfall, temperature) and global Brent oil prices. As a benchmark we estimate a naïve AR(1) model; we then build two regularised machine-learning specifications (Ridge regression): a short-term model for one-month-ahead prices using contemporaneous fundamentals and lagged prices, and a long-term model that predicts prices twelve months ahead using nine-month rolling averages of fundamentals and lagged prices. Out-of-sample results show that price inertia is strong: the naïve AR(1) achieves around R² ≈ 0.92 on the test set, whereas the short-term Ridge model performs worse (R² ≈ 0.79), indicating limited incremental value of richer fundamentals at very short horizons. In contrast, the long-term Ridge model delivers robust twelve-month-ahead forecasts (R² ≈ 0.90, MAE ≈ 6,200 VND/kg), capturing medium-run price dynamics and structural shifts as rolling macro-climate and trade conditions evolve. Building on the forecast distribution and error bands, we derive a “fair price corridor” around expected future prices, which can guide the timing of sales, provide transparent reference ranges for farm-gate negotiations, and inform the design of support policies for smallholders. The main contribution is to operationalise fair dynamic pricing in an agricultural emerging-market context by tightly coupling econometric–machine learning forecasts with fairness economics, offering a framework that can be embedded in digital advisory tools for coffee farmers and adapted to other perennial crops.

ABSTRACT. This study explores the role of digital simulation in Content-Based Instruction (CBI) for nutrition students, emphasizing how digital simulation from multimedia resources contributes to the teaching and learning process and how both lecturers and students perceive its use. Employing a qualitative approach, data were collected through classroom observations, semi-structured interviews, and supporting documentation. The findings indicate that digital simulation serves as an effective supplementary tool in fostering student engagement, comprehension, and contextual learning of English within nutrition-related topics. Students reported that the integration of digital simulation made lessons more interactive, easier to follow, and highly relevant to their field of study. Likewise, the lecturer perceived digital simulation as a valuable medium to bridge language learning with discipline-specific content, allowing for more authentic classroom experiences. Almost all interview participants expressed positive responses, highlighting increased motivation, improved vocabulary retention, and greater confidence in using English. The study concludes that digital simulation significantly enhances the implementation of CBI by supporting both linguistic and subject matter understanding. These findings suggest that incorporating digital simulation in English for Specific Purposes (ESP) courses can enrich students’ learning experiences and provide lecturers with creative avenues to deliver content more effectively. Keywords: Content-Based Instruction, Digital Simulation, Nutrition Students

ABSTRACT. The integration of artificial intelligence (AI) into hospitality and tourism has accelerated digital transformation, reshaping how organizations innovate, deliver services, and co-create experiences. Despite growing scholarly attention, research on AI’s role in hospitality service encounters remains fragmented, with limited synthesis of conceptual, methodological, and thematic developments. This study addresses this gap through a systematic literature review that combines bibliometric mapping and text-mining analysis to identify emerging trends, key themes, and research frontiers in AI-driven hospitality. Following PRISMA protocols, 487 peer-reviewed articles published between 2000 and 2024 were analyzed using co-occurrence networks, clustering algorithms, and TF-IDF keyword extraction. Six major thematic clusters were identified, representing focal areas such as service automation, customer experience, cultural adaptation, chatbot communication, and human–robot collaboration. The findings reveal a paradigm shift in the field's intellectual structure, moving decisively from efficiency and automation to a human-centric focus on empathy, ethics, and affective intelligence. This study contributes a holistic synthesis of the field, bridging technological and human-centered perspectives in hospitality. By mapping intellectual structures and conceptual gaps, it provides a foundation for future research agendas that align AI adoption with empathy, authenticity, and sustainable service design.

ABSTRACT. The integration of artificial intelligence (AI) and business analytics (BA) into accounting information systems (AIS) represents a paradigm shift from passive data processing to intelligent, autonomous decision support. This study conducts a systematic literature review of 58 peer-reviewed articles published between 2016 and 2025, utilizing the Scopus, ProQuest, and ScienceDirect databases to map the trajectory of this digital transformation. Grounded in the Service Science framework, we synthesize fragmented research into a coherent conceptual model that links technological enablers to organizational capabilities and performance outcomes. Our findings demonstrate that while AI-enabled AIS significantly enhances data accuracy, real-time forecasting, and fraud detection, it simultaneously introduces critical friction points regarding data governance, algorithmic transparency ("black box" issues), and ethical accountability. The study contributes to the literature by identifying the specific mechanisms through which AI reshapes auditing and management control, offering a roadmap for practitioners navigating the trade-offs between automation efficiency and professional judgment. Finally, we outline a future research agenda focused on the convergence of AI with blockchain and the emerging role of intelligent systems in ESG reporting and sustainability governance.

ABSTRACT. This study examines the relationship between firms’ environmental, social and governance performance and their disclosure of digital transformation in annual reports. Using a panel of 79 non-financial firms listed on the Australian Securities Exchange from 2014 to 2024, we construct a text-based digital transformation index and estimate two-way fixed effects models with Driscoll–Kraay robust standard errors. The results show a negative and statistically significant association between ESG performance and digital transformation disclosure, indicating that firms with stronger ESG commitment place less emphasis on communicating digital initiatives. This suggests a strategic substitution effect, where sustainability-oriented firms rely less on digital signaling in their corporate reporting. The study contributes to understanding how firms balance competing strategic domains in their disclosures, and provides insights for managers and policymakers regarding integrated approaches to sustainability and digital transformation disclosure.

ABSTRACT. Digital transformation has emerged as a critical driver of organizational agility for SMEs in the post-pandemic era, yet the measurement of agility and its predictors remains underexplored at scale. This study addresses this gap by leveraging a comprehensive, cross-national dataset of 491,636 SME technology adoption records, distinguishing between production and consumption technologies. Using advanced machine learning techniques—including Random Forest, XGBoost, and ensemble models—the research systematically evaluated the predictive power of technology adoption indicators on organizational agility outcomes. The Random Forest model achieved the highest test R² (0.6249), demonstrating superior ability to capture the complex, non-linear relationships that characterize SME digitalization and agility. Key findings reveal that production technologies, such as automation and digital infrastructure, are stronger predictors of agility than widely adopted consumption technologies, challenging prevailing assumptions about the primacy of customer-facing digital tools. These insights not only advance theoretical understanding by integrating the Technology-Organization-Environment and dynamic capability frameworks, but also offer actionable guidance for SME leaders and policymakers. The results underscore the need for targeted investment in foundational digital capabilities and workforce upskilling to foster resilience and adaptability. Ultimately, this study demonstrates the transformative potential of machine learning in guiding data-driven digitalization strategies, equipping SMEs to navigate uncertainty and drive sustainable recovery in a rapidly evolving business landscape.

ABSTRACT. As artificial intelligence (AI) becomes embedded in everyday communication, people increasingly engage with chatbots not only as tools but as companions. This study explores how Vietnamese users experience and interpret friendship with ChatGPT, a general-purpose AI not designed for emotional bonding. Using Brandtzaeg et al.’s (2022) seven friendship dimensions (voluntariness, reciprocity, intimacy, similarity, self-disclosure, empathy, and trust) and integrating the frameworks of Computer-Mediated Communication (CMC), Networked Individualism, and Computers Are Social Actors (CASA), this qualitative research analyzes 16 semi-structured interviews with active ChatGPT users in Vietnam. Findings reveal that human-AI friendship develops organically through sustained interaction rather than design intention. Users described their relationships with ChatGPT as voluntary, self-paced, and functionally reciprocal, valuing autonomy, safety, and consistency over emotional authenticity. ChatGPT’s responsiveness and linguistic empathy fostered comfort and trust, leading participants to attribute social meaning to non-human communication. Yet, users remained aware of its artificial nature, negotiating authenticity through conscious anthropomorphism. The study contributes to emerging theories of human-AI companionship by conceptualizing friendship as a processual and reflexive sociotechnical relationship, one co-produced through language, interpretation, and user agency. It highlights a broader cultural shift in digital sociality, where connection, care, and trust are increasingly distributed between humans and intelligent machines.

ABSTRACT. This study examines the earnings gap between formal and informal workers in Vietnam using microdata from the 2021 Labour Force Survey, a period marked by substantial labour market disruption due to Covid-19. A baseline Blinder–Oaxaca decomposition is first employed to distinguish the portion of the gap attributable to observable characteristics from the portion driven by differences in returns, revealing that structural wage-setting factors dominate while compositional differences play only a limited role. To extend the analysis beyond the linear Oaxaca framework, a supervised machine-learning model is used to predict individual incomes, and SHAP values are applied to decompose the predicted income gap. The ML–SHAP approach confirms the importance of education, occupation and regional characteristics, while uncovering nonlinear and interaction effects that are not captured by traditional econometric methods. Counterfactual simulations further illustrate how informal workers’ earnings would change under alternative distributions of education, region and enterprise type. Overall, the results demonstrate that machine-learning methods can validate and enrich conventional decomposition analysis, offering a more comprehensive empirical foundation for designing targeted policies to improve income prospects for workers in the informal sector.

ABSTRACT. An efficient use of water is crucial in farming sustainability, in areas that have limited water. Fixed schedules or manual judgment, which is the basis of the traditional irrigation systems, usually lead to an over-or under-irrigation that influences the yields and waste resources. This paper suggests a smart irrigation system that can be performed through a machine learning system with the help of sensor data to make ideal irrigation choices. Six monitored algorithms, logistic regression, random forest, gradient boosting, AdaBoost, support vector machine, K-nearest neighbors machine learning related 2000 samples (irrigation_machine.csv) measured over three agricultural plots with 20 sensor characteristics and irrigation answers of either 0 or 1. Normalization, one-hot encodings and train-test split 70-30 were employed in preprocessing. The accuracy, precision, recall, and F1-score were used to measure performance. The best results were achieved by AdaBoost where 99.45 was recorded as the accuracy, 99.45 was the precision, 99.47 was the recall, and 99.52 was the F1- score. The ability to get feature relevance and class balance insight by using correlation heatmaps was observed. This could be done in addition to making accuracy and efficiency more enhanced although there was no weather data. The suggested AI-based system effectively provides a high level of predictive performance, which minimizes the amount of unused water and promotes efficient precision irrigation at any scale and in any agricultural environment.

ABSTRACT. Achieving the Sustainable Development Goals (SDGs) poses a distinct challenge for large companies, especially concerning Goal 13 on climate change mitigation. One of the key challenges in meeting this goal is addressing Scope 3 emissions, which make up more than 90% of corporate emissions. However, dealing with these emissions is challenging due to the intricate web of upstream and downstream suppliers from which the emission data must be collected. This paper presents a novel approach using domain-adapted Natural Language Processing (NLP) foundation models to estimate Scope 3 emissions by automating transaction classification into commodity classes using financial transaction data. The approach is evaluated against text classification benchmarks, TF-IDF, Word2Vec, and zero-shot learning. Our results indicate that domain-adapted models significantly outperform classical techniques, as evidenced by the Word2Vec model achieving a higher F1 score of 72% compared to 69% for TF-IDF. Zero-shot classification based on commodity descriptions yielded better results than using titles, producing F1 scores from 40.1% to 43.7%. A fine-tuned Roberta-base model achieved the highest F1 score of 87.2%, improving to 87.19% with a lower learning rate. These results show the effectiveness of automating Scope 3 emissions estimation at an enterprise level, enabling proactive climate action in alignment with SDG 13 for reduced carbon emissions.

ABSTRACT. Agricultural sustainability faces mounting challenges from climate change, soil degradation, and the growing demand for food production. Crop rotation has long been recognized as an effective practice for enhancing soil fertility, mitigating pest infestations, and boosting yields. However, traditional rotation planning relies heavily on farmer experience and static guidelines, which often fail to account for dynamic variables such as climate variability, market conditions, and evolving soil health profiles. Recent advances in Artificial Intelligence (AI) and predictive analytics offer a transformative approach to optimizing these decisions through data-driven modeling. This study proposes a hybrid Convolutional Long Short-Term Memory (ConvLSTM) model that integrates Conv1D layers for short-term temporal feature extraction with LSTM layers for long-term dependency modeling. The methodology utilizes multi-year agricultural time-series data from the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT), encompassing production, yield, and cultivated area across key crop types. Experimental results demonstrate that the proposed ConvLSTM achieves superior classification performance, attaining an overall accuracy of 94.8%, outperforming traditional LSTM, GRU, and machine learning baselines. Class-wise evaluations reveal particularly high precision and recall in major crops such as rice, wheat, and pearl millet, alongside stable convergence patterns with minimal overfitting. These findings highlight the potential of ConvLSTM-based predictive analytics to revolutionize crop rotation planning by providing scalable, accurate, and adaptive recommendations. Such systems could serve as essential tools for promoting sustainable agriculture, enhancing productivity, and improving resilience to climate-induced disruptions.

ABSTRACT. This study investigates the influence of the digital economy on green growth across 39 Asian countries during the period 2010 to 2022. By employing panel data techniques and estimating Ordinary Least Squares, Fixed Effects, and Random Effects models, the analysis identifies the Random Effects specification as the most appropriate based on Breusch Pagan LM and Hausman tests. The findings indicate that the digital economy serves as a significant driver of green growth in Asia, reinforcing the transformative role of digital technologies in enhancing resource efficiency and reducing environmental pressures. Additional results reveal that foreign direct investment, government expenditure, and regulatory quality contribute positively to green growth, while economic development and financial development show negative effects. Furthermore, the interaction between the digital economy and government regulations demonstrates a strong amplifying effect, suggesting that regulatory frameworks can guide digital transformation toward environmentally sustainable outcomes. The study underscores the need for coherent policy strategies that integrate digital innovation with green development objectives. These include strengthening green technology adoption, enhancing regulatory quality, expanding green finance mechanisms, promoting sustainable trade practices, and accelerating smart urban development. The results provide valuable implications for policymakers seeking to leverage digital transformation as a pathway to sustainable and low carbon growth in the Asian region.

ABSTRACT. The long-standing economic disparities between the Global North and South testify to the inadequacy and inequity of the present global economic governance (GEG). Since the mid-1940s, the United States and other major industrial economies maintained the North-dominated international economic order. Later, these powerful northern nations established GEG mechanisms including the G7 and the G20 to retain their dominating roles. In keeping with the growing influence of developing countries over the global economy and world politics since the 1960s, many southern nations also created their GEG mechanisms including the G77 and the BRICS to participate in the GEG system. This article examines GEG in the context of the Global North-South divide. It provides a historical overview of postwar GEG development by summarising the emergence of governance actors at the national, institutional and global levels. Importantly, the article calls for the building of an equitable international economic order to bridge the North-South divide through strengthening cooperative GEG mechanisms. In an increasingly multipolar world, scholars, policymakers and practitioners need to examine further the potential pivotal role of middle and regional powers in reforming the present global governance system and creating an alternative international economic order, thus tackling the growing economic and geopolitical fragmentation.

ABSTRACT. This study develops and tests an integrated theoretical model named Trust Innovation to examine how key determinants—namely Sharia compliance, transparency, and security—influence the intention to adopt Sharia fintech through the simultaneous mediating roles of trust and perceived value. Utilizing a quantitative approach, data were collected from 400 active Sharia fintech users across Indonesia aged 17 years and above. Nine proposed hypotheses were empirically analyzed using Partial Least Squares (PLS)-based Structural Equation Modeling (SEM). The results confirm that the three antecedent factors significantly affect both trust and perceived value, affirming that user perceptions are grounded in ethical assurances and functional performance. Crucially, trust demonstrated a stronger direct effect on adoption intention (H9) compared to perceived value (H8), highlighting its dominant role within the Sharia context. Additionally, findings indicate that perceived value significantly enhances user trust (H7), suggesting that initial utility contributes to deeper conviction. This study offers important implications for theory development, practice, and policy in Sharia fintech in Indonesia. Theoretically, the Trust Innovation model broadens technology adoption understanding by incorporating both technical and ethical-religious dimensions. Sharia compliance, transparency, and security are shown to shape user trust and perceived value, indicating that moral legitimacy is a primary factor alongside functional benefits. Practically, service providers need to strengthen Sharia compliance and digital security aspects, while regulators should promote Sharia audits and Islamic value-based financial literacy. Keywords: Sharia Fintech; Trust; Sharia Compliance; Perceived Value; Adoption Intention; Financial Innovation.

ABSTRACT. Purpose: To explain how entrepreneurial orientation (EO) helps small and medium-sized enterprises (SMEs) progress along a four-level sustainability maturity ladder, distinguishing environmental and social (HR & community) practices, and to test whether leaders’ biospheric and altruistic values strengthen this relationship. Design/methodology/approach: We analyze survey data from 409 canadian SMEs. We construct maturity indices for environmental and social practices and classify firms into four stages (lowest → highest). We estimate ordered logit and probit models with robust checks, including controls for firm size (log employment), cooperative status, premises ownership, and regional dummies. We report odds ratios and average marginal effects (AMEs) with parametric confidence intervals and verify proportional-odds assumptions and out-of-sample calibration heuristics. Research limitations/implications: Cross-sectional, self-reported data limit causal inference and raise common-method concerns. The maturity staging is survey-based and may vary across contexts. Results come from one dataset and should be replicated with longitudinal or multi-source designs and complemented with objective performance outcomes. Practical implications: SMEs with stronger EO are more likely to be at higher sustainability stages, especially for people-focused practices. A one-SD increase in EO is associated with roughly 62–79% higher odds of being in a higher stage and +7–9 percentage-points in the chance of reaching the top stage (env/social models, with controls). Managers can accelerate maturity by nurturing EO capabilities (small-scale experimentation, proactive scanning, disciplined risk management) while institutionalizing HR and community practices. Social implications: Because EO most strongly supports social practices, policies and programs that develop entrepreneurial capabilities in SMEs (training, mentoring, peer learning) can yield broader social benefits, better workplaces and deeper community engagement, alongside environmental improvements. Originality/value: We integrate maturity modeling with capability-based strategy, separating environmental and social ladders, and we test value-based moderation. We find consistent main effects of EO and positive main effects of biospheric/altruistic values, but no systematic moderation, clarifying when values matter in sustainability upgrading.

ABSTRACT. Climate change mitigation depends on accurate accounting of greenhouse gas (GHG) emissions, yet Scope 3 emissions, which arise from indirect supply chain activities, remain largely underreported de- spite constituting the majority of many firms’ carbon footprints. Un- derstanding how to estimate and manage these emissions is essential for achieving national decarbonization goals and corporate sustainabil- ity commitments. Although numerous macro-level methodologies exist to estimate Scope 3 emissions, few leverage multi-dimensional datasets that jointly capture value added, employmen t, energy use, and GHG emissions across all NAICS sectors in the U.S. Existing computational approaches also often fail to model the interdependent and nonlinear re- lationships driving carbon intensity. To address these gaps, this study develops a hybrid machine learning framework that integrates Princi- pal Component Analysis for dimensionality reduction, Gaussian Mixture Modeling for uncovering latent emission clusters, and Natural Gradient Boosting for uncertainty-aware predictive regression. Using harmonized firm-level datasets mapped across NAICS and USEEIO codes, the model identifies complex structural relationships among operational, environ- mental, and economic variables influencing Scope 3 emissions across U.S. industries. Model performance, evaluated using Root Mean Square Error, Mean Absolute Error, and the coefficient of determination, shows sub- stantial accuracy gains compared with ensemble baselines such as Cat- Boost, Random Forest, and Gradient Boosting. The analysis also reveals interpretable emission clusters that highlight variations in firm behavior and supply chain structures. These findings provide accurate predictive carbon intensity for U.S. enterprises and demonstrate how data-driven clustering and regression can enhance transparency, regulatory compli- ance, and evidence-based decarbonization strategies.

ABSTRACT. This study explores the role of export strategies in overcoming the middle-income trap, focusing on the experiences of Asian economies that have successfully transitioned to high-income status. Using a comparative analysis of seven countries - Japan, South Korea, Malaysia, the Philippines, China, Indonesia, and Thailand - this research examines the export structures and growth patterns that have enabled some nations to escape the middle-income trap while others remain ensnared. The findings highlight the importance of high-tech manufacturing, export diversification, and participation in global value chains as key drivers of economic advancement. Additionally, the study emphasizes the growing significance of digital and high-value service exports in the era of globalization. Drawing on these insights, the paper offers strategic recommendations for developing economies, such as Vietnam, to enhance their export models, diversify their markets, and pursue deeper integration into global trade networks. The results provide a roadmap for nations seeking to accelerate their path to high-income status and achieve sustainable economic growth.

ABSTRACT. The study examines how university students in the UAE perceive climate change and investigates what factors influence these perceptions by conducting a questionnaire survey of undergraduate students from the University of Wollongong in Dubai and the Canadian University Dubai. Results indicate that students’ perceptions of climate change are shaped primarily by their gender, nationality, and awareness of climate issues. Female students are found to exhibit greater urgency, motivation, and concern about climate change, while students from the MENA region display comparatively lower engagement. Moreover, students who are more informed about climate change tend to express stronger urgency, increased motivation to act, and greater concern over its worsening impact. These findings provide a deeper understanding of climate-related awareness among students and offer valuable information to design effective policies and interventions to enhance climate engagement among university students in the UAE, as well as among youth in the region.

ABSTRACT. This study investigates how innovation culture and transformational leadership affect organizational performance in Vietnam’s real estate sector. Building on the resource based view and dynamic capabilities, an integrated model linking organizational resources, Technological Capabilities, innovation culture, transformational leadership, and organizational performance is tested. Survey data from 231 real estate firms were analyzed using PLS-SEM. The findings show that resources (β = 0.476, p < 0.001) and technology (β = 0.139, p < 0.05) strengthen innovation culture; innovation culture improves organizational performance (β = 0.362, p < 0.001); and transformational leadership directly enhances performance (β = 0.183, p < 0.001) but does not moderate the innovation performance relationship. Our findings add evidence for an industry that has been overlooked real e Corresponding author state. The results also show that leadership works differently depending on context, and suggest practical lessons for how firms in emerging economies can manage resources, apply technology, and develop leadership capacity

ABSTRACT. The rapid evolution of e-commerce has transformed global retail dynamics, intensifying competition and reshaping how consumers perceive value and risk in online transactions. Despite extensive research on service quality and digital marketing, the combined effects of personalized marketing (PM) and external cues on consumer perceptions remain underexplored. Grounded in the Stimulus–Organism–Response (S–O–R) framework, this study investigates how PM and external cues (e.g., brand reputation, pricing transparency, and interface design) influence customer-perceived value through the mediating mechanisms of service quality and perceived risk. Using a quantitative approach, data were collected from 564 Vietnamese online shoppers through a structured questionnaire and analyzed using Structural Equation Modeling (SEM) via LISREL. The findings reveal that PM exerts a significant positive effect on service quality (β = 0.23, t = 2.61) and a negative effect on perceived risk (β = –0.63, t = –7.15), indicating that tailored marketing strategies enhance trust and reduce uncertainty. Service quality demonstrates a strong positive impact on perceived value (β = 0.69, t = 10.3) and a negative influence on perceived risk (β = –0.17, t = –2.53), confirming its central role in shaping consumer confidence. External cues similarly enhance service quality (β = 0.76, t = 8.26) and mitigate perceived risk (β = –0.28, t = –3.04), illustrating that consistent branding and transparent communication can improve perceptions of reliability and authenticity. Theoretically, this study extends the S–O–R framework by integrating PM and external cues as dual stimuli that jointly shape consumer evaluations and value perceptions in digital marketplaces. Practically, it provides actionable insights for e-commerce managers seeking to enhance customer trust, engagement, and loyalty through data-driven personalization, quality assurance, and risk reduction strategies. By emphasizing the mediating roles of service quality and perceived risk, the study contributes to a more holistic understanding of how marketing stimuli translate into perceived value within contemporary e-commerce ecosystems.

ABSTRACT. In the era of booming sustainable digital transformation, organizations' approach must bridge the gap between internal and external marketing to drive sustainable brand development. This study adopts a human-centered perspective to analyze how internal marketing (IM) practices enhance employee engagement (EE) within an organization, which then drives brand advocacy (BA) and contributes to a more sustainable brand culture (SBC), serving as a critical external marketing factor amid the digital shift. Using Social Exchange Theory as the conceptual foundation, the suggested model was tested using the data from a 97-employee survey of a garment enterprise that has nurtured a human-intensive approach and is undergoing digital transformation in Vietnam. This is validated using the crucial role of employees as assets in aligning internal values with external brand articulation. The measurement model assessment first establishes reliability and validity, which includes discriminant validity via the HTMT test. Then, the structural analysis revealed a specific path in which IM functions as an influential driver of both EE (β = 0.742, p < .001) and BA (β = 0.606, p < .001). However, contrary to traditional beliefs, EE did not show a strong association with BA (β = 0.119, p > .05). On the contrary, BA appeared to be the dominant and positive mediator (β = 0.832, p<.001), explaining a substantial variance in SBC (R² = 0.756). The results also challenge the assumption that engagement is a prerequisite for advocacy, showing that in a specific blue-collar environment, tangible organizational support (IM) is the primary driver of Brand Advocacy, overriding the influence of psychological engagement. Moreover, the study also highlights that digital sustainability depends not only on technological advances but also on human-centered strategies that facilitate employees' brand-supportive behaviors. By determining the explicit link between IM and BA, the study aligns internal and external marketing under a cooperative, people-centered framework, which provides actionable insights for organizations whose mottos are to build humanized, resilient, value-driven brand cultures, a modern concept rooted in human-centric Industry 5.0.

ABSTRACT. Facing a paucity of research, this study was designed to investigate the interrelationships among AI knowledge (as a future work skill), AI self-efficacy, and career sustainability. Its main objective was to explore the mediating function of AI self-efficacy in the relationship between AI knowledge and career sustainability. Utilizing a cross-sectional design, data were conveniently sampled from 402 students at three Vietnamese universities via questionnaires. The hypotheses were empirically tested using Structural Equation Modeling (SEM). The results confirmed significant positive effects of AI knowledge on AI self-efficacy and career sustainability. A significant positive association between AI self-efficacy and career sustainability was also supported. Crucially, the analysis established a partial mediating role for AI self-efficacy. This research uniquely extends current knowledge on the role of AI competence in shaping higher education and professional development trajectories towards digital sustainability.

ABSTRACT. Abstract Purpose: This paper explores the pivotal role of digital leadership (DL) in fostering sustainable human resource management (SHRM) within emerging Asian enterprises. It aims to identify how leadership in the digital era enables organizations to balance technological transformation with sustainable people management and long-term performance. Design/methodology/approach: This qualitative–conceptual study adopts an integrative literature review and cross-case analysis of selected enterprises from emerging Asian economies. By synthesizing theoretical insights from the resource-based view (RBV) and stakeholder theory, the paper develops a conceptual framework linking digital leadership, sustainable HRM, and organizational performance. Findings: The study highlights that digital leadership practices—characterized by agility, innovation orientation, and data-driven decision-making—positively influence sustainable HRM dimensions such as employee engagement, green competence development, and digital well-being. The alignment of digital transformation and sustainability principles strengthens organizational resilience and enhances performance outcomes. Practical implications: The findings provide actionable insights for leaders seeking to integrate digital tools and sustainability values into HRM systems. Managers in emerging markets can leverage digital leadership capabilities to foster human-centered transformation and promote organizational agility. Originality/value: This study contributes to the growing body of knowledge by proposing a comprehensive framework linking digital leadership to sustainable HRM practices and organizational outcomes, contextualized within emerging Asian enterprises. It bridges a gap between digitalization and sustainability discourses in human capital management.

ABSTRACT. 123

ABSTRACT. The rapid growth of Artificial Intelligence (AI) has significantly transformed the education globally, and Vietnam’s English centers is now operating an important phase of technological transition. This research paper examines how AI has reformed English centers in Ho Chi Minh City (HCMC) - one of Vietnam’s largest academic hub - shifting from conventional, teacher-based classroom to AI-enhanced learning environment. Applying a mixed-methods approach - comprising surveys with 200 learners and 50 educators, structured interviews with administrators across three pioneering English institutions - the study analyses the multifaceted of AI applications such as adaptive learning platforms, automated feedback systems, and traditional chatbots. Findings show that AI significantly boosts student engagement, efficiency, and personalization. However, the research also identifies notable challenges related to technological infrastructure, cost effectiveness, teacher readiness and the limit of human interaction in language learning. The paper concludes that although AI cannot replace teachers, it serves as a catalyst for pedagogical and operational innovation, empowering English centers in HCMC to enhance accessibility and strengthen competitiveness within an increasingly digital educational environment. Recommendations focus on ensuring that AI integration in Vietnam’s English education sector is sustainable, ethical, and inclusive.

ABSTRACT. Purpose - This study examines how Construction 4.0 (C4.0) technologies influence the sustainable performance of Vietnamese construction firms, focusing on the post-adoption phase. It investigates how five key attributes of technology adoption - relative advantage, compatibility, trialability, observability, and complexity - affect economic, social, and environmental dimensions of sustainability.

Design/methodology/approach - Grounded in the Diffusion of Innovation (DOI) and Socio-technical Systems (STS) theories, this study adopts a quantitative approach using survey data from 215 construction firms in Vietnam’s Mekong Delta. Structural Equation Modeling (SEM) was employed to analyze the impacts of the five C4.0 attributes on sustainable performance, including their mediating relationships.

Findings - The findings reveal that all five attributes significantly enhance sustainable performance, with compatibility, trialability, observability, and complexity mediating the relationship between relative advantage and sustainability outcomes. These results highlight the critical role of leveraging C4.0 technologies in addressing sustainability challenges through enhanced workflows, resource optimization, and integration of technological and social dimensions in construction practices.

Originality/value - This research provides novel insights into the post-adoption impacts of C4.0 technologies in a developing country context, addressing the gap in empirical studies on sustainable performance. It underscores the importance of aligning C4.0 technology attributes with organizational processes to achieve triple bottom-line sustainability in construction firms.

ABSTRACT. Export-oriented garment supply chains in emerging economies are exposed to intensifying sustainability-related supply chain risks (SSCRs), including economic shocks, social and labour compliance pressure, and environmental scrutiny. At the same time, global buyers and regulators expect reliability, transparency, and responsible conduct. Industry 4.0 technologies (referred to here as Tech4.0) are often promoted as solutions for visibility, traceability, and control. However, there is still limited empirical evidence on whether technology-enabled risk management actually protects supply chain performance (SCP) when suppliers operate under sustainability pressure in labour-intensive, multi-tier garment networks. This study addresses that gap by treating Tech4.0 not as a generic digital upgrade, but as an operational risk management capability. The paper is guided by the Resource-Based View and Stakeholder Theory. SSCRs are modelled across three dimensions (economic, social, and environmental) as predictors of three SCP outcomes, including the supplier performance, internal process performance, and customer-facing performance. The analysis then tests whether Tech4.0 capabilities such as real-time sensing, digital traceability, predictive planning, and data-driven coordination moderate these relationships. The model is estimated using variance-based structural equation modelling (PLS-SEM) on survey data from 275 Vietnamese garment firms that operate in global supply networks. Vietnam is a strategically important export base for branded apparel, so the ability to manage sustainability pressure without performance loss is commercially decisive. Preliminary findings indicate that economic and social risks are positively associated with reported performance, which suggests that firms respond to these pressures by activating internal capabilities. Environmental risk shows no direct performance effect. Most importantly, Tech4.0 adoption does not automatically neutralize SSCRs. Its moderating effect depends on whether digital tools are embedded in formal risk routines and supported by workforce competence, rather than being deployed in isolation. The study contributes new evidence from an emerging market context, clarifies when technology actually delivers resilience, and offers guidance on how to align digital investment with sustainability-driven risk management.

ABSTRACT. Gold price forecasting plays a crucial role in financial decision - making and risk management. In this study, we develop and evaluate a Kalman filter-based state estimation framework for short-term prediction of gold prices. The proposed model treats the price evolution as a stochastic dynamic system, where hidden states capture the underlying market trend and observation noise represents market volatility. In the empirical analysis, we forecast gold prices using both historical data of world gold price and historical data of world gold prices influenced by key financial-economic variables, including U.S. Federal Funds Effective Interest Rate, U.S. Consumer Price Index, U.S. Dollar Index, Crude Oil Price and S&P 500 Index. Results confirm that the Kalman filter provides a robust and interpretable tool for gold price estimation, with potential for integration into hybrid models combining Kalman filter and deep learning architectures. Specifically, we explore approaches Kalman Filter with Long Short-Term Memory (KF-LSTM) and Kalman Filter with Gated Recurrent Units (KF-GRU, both of which integrate neural networks into the state-transition mechanism. Comparative experiments demonstrate that the proposed hybrid models achieves superior forecasting accuracy, particularly in during periods of heightened market volatility.

ABSTRACT. The primary objective of this research was to empirically assess the relationship between social capital and the entrepreneurial intentions and decisions of economics students enrolled at universities in Ho Chi Minh City, a context increasingly defined by the pervasive societal influence of digital transformation. This digital shift not only fosters new business models but fundamentally alters the formation, maintenance and leveraging of social capital itself. Primary data were systematically collected through a survey administered to a robust sample of 400 students, ensuring a solid basis for subsequent statistical inference. The methodology employed a comprehensive analytical sequence. This included utilizing Cronbach's Alpha for scale reliability, followed by Exploratory Factor Analysis (EFA) and subsequently Confirmatory Factor Analysis (CFA) to validate the measurement model, and finally, Structural Equation Modeling (SEM) to rigorously test the causal relationships within the proposed theoretical framework. The key findings strongly indicated that various dimensions of social capital—specifically cognitive capital of aspiration, perceived feasibility capital, relational capital, structural capital, and financial capital—all exerted a significant and positive influence on students' propensity to engage in entrepreneurial activity. Notably, in the digital era, relational and structural capital are increasingly mediated and amplified through online social platforms and connectivity tools, thereby enhancing access to critical entrepreneurial information and resources. Furthermore, the study identified the university educational environment as a crucial moderating variable, which amplified the positive impact of social capital determinants on students' entrepreneurial intentions. Based on these empirical results, the research concludes by proposing specific managerial and policy recommendations aimed at cultivating social capital and thereby enhancing entrepreneurial activities among economics students within the rapidly evolving Ho Chi Minh City educational landscape.

ABSTRACT. This study investigates the psychological and contextual factors shaping Vietnamese university students’ intention to participate in international internship programs. Our study based on Theory of Planned Behavior (TPB) theoretical framework including antecedents - financial burden, policy support, and language confidence and the core constructs - attitude, subjective norms, and perceived behavioral control. A structured survey will be administered to undergraduates at multiple institutions in Vietnam. Data will be analyzed using Structural Equation Modeling (SEM) in R, with procedural and statistical controls applied to mitigate common method bias. The model aims to clarify indirect pathways linking structural barriers to behavioral intention, providing both theoretical refinement and practical guidance for universities and policymakers seeking to expand equitable access to global internships.

ABSTRACT. This study evaluates whether ensemble machine learning models can deliver economically meaningful forecasts for daily S&P 500 returns over a quarter-century horizon. Using 6,266 trading day observations (January 2000–May 2024) drawn from CRSP, we engineer eight technical predictors capturing short-term momentum, volatility clustering, and micro structure noise. A Random Forest (200 trees) and Gradient Boosting regressor (400 stages) are trained on the 2000 2019 window; a convex blend (60 % RF, 40 % GB) forms the ensemble. Hold out testing on the pandemic era 2020 2024 sample shows the ensemble trimming mean absolute error by 3 % relative to its constituents, yet failing to outperform a naïve zero return benchmark in R^2. A long-only trading rule conditioned on positive forecasts reduces maximum draw down from 33.9 % to 20.5 % and lowers annualised volatility by six percentage points but yields a modest 2.1 % CAGR, insufficient after transaction costs. Permutation importance analysis attributes 39 % of predictive power to MACD momentum and 21-day volatility, underscoring the dominance of behavioural factors at the daily horizon. Year by year error diagnostics reveal acute regime sensitivity, with performance deteriorating during the COVID-19 crash and 2022 inflationary sell-off. We conclude that while transparent tree ensembles enhance risk management, achieving consistent out performance necessitates regime aware or deep sequence models augmented with macro sentiment signals.

ABSTRACT. The rapid development of Artificial Intelligence (AI) has posed profound legal challenges to the protection of personal data. AI systems often collect, process, and analyze large volumes of personal information to train models, make automated decisions, and personalize services. This paper conducts a qualitative comparative legal analysis of Vietnam’s data protection framework under Decree No. 13/2023/NĐ-CP and the European Union’s General Data Protection Regulation (GDPR). It examines similarities and differences in four key areas: principles of data processing, rights of data subjects, obligations of data controllers, and enforcement mechanisms. The study highlights the degree of convergence between Vietnam’s regulations and international standards, identifies existing legal gaps in governing AI-driven data processing, and proposes recommendations for aligning Vietnam’s framework with global best practices while ensuring citizens’ privacy rights.