Earticle

Cold chain is crucial to ensure product safety and quality, but the risk is higher than tradi-tional supply chain due to the complexity of the operation, such as the need for strict temper-ature control and reliance on specialized equipment. Therefore, implementing strategic measures to maintain temperature control, product quality, and safety while effectively managing and carefully monitoring these risk factors are essential components of cold chain management. This study focuses on exploring the potential hazards inherent in cold chain processes for perishable food and medicine products, which directly affect the quality of human life. We used web crawling techniques to meticulously collect from Google News articles that are related to cold chain risks. Using data collected from the Google News platform from January 2015 to April 2022, we leveraged latent Dirichlet allocation (LDA) topic modeling to systematically extract risk factors in the fresh food and medicine cold chains domain. We then calculated the importance of each topic based on word frequency probabilities. In doing so, we comparatively analyzed the differences and similarities of cold chain elements in the two industries and empirically validated them using non-parametric statistical techniques. The results of this study provide important insights for understanding and mitigating supply chain risks within the cold chain industry.

The field of Information Systems (IS) has emerged as a core domain within the convergence paradigm that integrates information technology (IT) and management to advance both academic and practical applications. As developments in information and communication technology (ICT) have made large-scale data more accessible, research efforts have increasingly focused on uncovering the structural evolution of IS knowledge systems and research trends. Previous studies have employed text mining and network analysis to analyze titles, abstracts, and keywords from IS journals; however, most have been limited to specific time periods or domestic journals, lacking a longitudinal and comparative perspective across leading global and regional journals. To address this gap, this study collected all articles published in the Asia Pacific Journal of Information Systems (APJIS) (1991–Q3 2022) and Management Information Systems Quarterly (MISQ) (1977–Q3 2022). The extracted keywords were categorized into research topics, methodologies, and theories, followed by four analytical approaches: frequency analysis, trend analysis, topic modeling, and keyword network analysis. These methods were used to identify and compare the longitudinal research trends and intellectual structures of the two journals. The findings reveal that both APJIS and MISQ have evolved from system- and technology-centered studies toward data-driven, AI-enabled, and user-focused research. While MISQ has consistently led the introduction of new research themes and theoretical frameworks, APJIS has expanded and contextualized these themes within the Asia-Pacific setting. This study is expected to provide practical guidance for IS journal stakeholders and future researchers by identifying the developmental trajectories of global and regional IS research and proposing promising directions for future studies.

Understanding what makes online reviews helpful remains a central concern in information systems research and platform design. While prior studies have examined sentiment and structure, less is known about how distinct linguistic tones―analytical, authentic, and mixed emotional―signal reviewer credibility in text-only environments. Drawing on Source Credibility Theory (SCT) and Cognitive Fit Theory (CFT), we propose that these tones act as stylistic heuristics for perceived expertise and trustworthiness, particularly when identity cues are absent and tone aligns with the reader’s evaluative goals. Using LIWC to analyze 54,759 Amazon reviews across search and experience goods, we find that all three tones enhance perceived helpfulness, but their effects vary by product type. Analytical tone is more persuasive for search goods, where structured reasoning signals expertise and supports attribute-based comparisons. Authentic tone is more effective for experience goods, where sincerity supports trust and affective simulation. We also introduce mixed emotional tone―the co-expression of positive and negative affect―as a novel signal of deliberation, which improves helpfulness ratings, especially for search goods. These findings clarify prior inconsistencies in sentiment research and advance a theory-driven framework for how tone–product fit influences review helpfulness in digital platforms.

As a new trend, live commerce offers a unique ecosystem within digital platforms reshaping consumer interactions that facilitates real-time product evaluation and informed decision-making. The theoretical lens, grounded in the S-O-R framework and involvement theory, investigates the interactivity factors of the live-streamer and the live platform. The data was collected from 448 live commerce users to empirically test the conceptual model. Using hybrid SEM-ANN analysis, the findings reveal that the perceived similarity, expertise, and familiarity of the streamer, as well as synchronicity & two-way communication, positively influence consumers’ cognitive and affective involvement. Specifically, perceived similarity showed the strongest influence on involvement. However, the relationship between active control and consumer involvement couldn’t be established. In particular, affective involvement exerted a relatively stronger impact on purchase intention than either flow or cognitive involvement. The findings offer insights on essential strategic decisions for e-retail platform investing in live commerce.

Digital platforms have gained prominence across various domains and industries by offering generativity and flexibility. In contrast to a vast number of benefits brought by digital platforms, the growing problem of digital debt accrual while embedding a digital platform into an organization’s digital infrastructure and associated work processes has attracted increasing organizational concern and recent academic attention. Drawing on technology affordance theory and paradox theory, this paper examines how the affordances and constraints of a digital platform are leveraged to contribute to gradual digital transformation. The findings of this in-depth case study identify different types of affordances that foster digital transformation: minimum viable product development, cloud-based collaboration, and low-code/no-code development. This study also identifies the constraints of a digital platform, which are different types of digital debt: design debt, capability debt, and technical debt, that may hinder digital transformation. This study discusses the paradoxical tensions that arise from the affordances and constraints of a digital platform and the potential paradox management strategies to navigate those tensions and achieve gradual digital transformation. This research provides insights for scholars, managers, and organizations to better understand the dynamics of leveraging digital platforms in facilitating digital transformation.

The rapid expansion of generative AI and cloud computing is transforming industrial structures and driving unprecedented growth in data centers. As a result, electricity demand is becoming increasingly shaped by digital technologies rather than traditional climate or economic factors alone. However, existing forecasting models largely overlook these sociotechnical drivers, risking substantial underestimation of future demand. This study integrates technology diffusion indicators with climate variables to examine whether digital-technology trends meaningfully contribute to national electricity demand. Using a Double Machine Learning framework as a feature-validation step, we confirm that GPT search volume and cloud market size serve as statistically robust and predictive indicators of electricity consumption. In addition, Fourier Transform–based features are employed to capture periodic variability, significantly improving forecasting performance beyond climate- and economy-centric baselines. Scenario simulations under the SSP585 pathway forecast a steady rise in demand through 2045, with seasonal peaks amplified by AI and cloud adoption. The findings highlight the structural role of sociotechnical factors in shaping electricity demand and offer practical implications for electricity pricing, infrastructure planning, and risk management.

With growing concerns about digital privacy, understanding the mechanisms that motivate or hinder privacy-protective behavior has become increasingly critical yet challenging. While Protection Motivation Theory (PMT) traditionally emphasizes motivational determinants, this study highlights the often-overlooked role of habitual resistance. Drawing upon PMT, we propose a dual-pathway model that simultaneously examines factors that both encourage and disrupt privacy-protective behavior. Our proposed extended PMT model was empirically tested via an online survey targeting Korean internet users, yielding 309 valid responses. We used SPSS 29.0.2.0 to obtain descriptive statistics and SmartPLS 4.1.0.3 to estimate the structural equation model. Findings revealed that both perceived threats and perceived efficacy significantly enhanced motivation, which in turn strongly predicted privacy-protective behavior. However, maladaptive habits negatively affected protective behavior, even when motivation was strong. Interestingly, perceived efficacy did not significantly reduce maladaptive habits, indicating that motivation alone was insufficient to change established behavior patterns. Our study advances online information privacy research by synthesizing motivational drivers and habitual barriers within a unified framework. The findings call for more comprehensive interventions that address often underestimated automatic behavioral tendencies. This study also offers insights for designing privacy education and policy tools that target unintentional barriers to protective action.

The exponential growth of online information has posed significant challenges for analyzing and classifying user sentiments, especially in complex linguistic contexts. This paper introduces a novel approach combining Ontology and a fine-tuned Transformer model (BERT) for automated sentiment analysis. Unlike traditional methods, our approach integrates a domain-specific Ontology to enhance contextual understanding and semantic reasoning, addressing challenges like sarcasm and ambiguous expressions. Experimental results on diverse datasets, including Vietnamese social media comments and international sentiment datasets, demonstrate the superiority of our system, achieving a 95.64% Quality of Experience (QoE) score―significantly outperforming existing methods. This study not only advances sentiment analysis techniques but also provides insights into improving user satisfaction across various applications, including customer feedback analysis, education, and public opinion monitoring. Our results indicate higher label-wise precision (QoE) on Vietnamese social comments; we also provide brief qualitative examples illustrating when the ontology helps.

Cloud-native technology has recently gained significant attention as an enterprise application capable of overcoming the limitations of traditional on-premise systems and delivering greater value. By leveraging cloud-based infrastructure, organizations can achieve scalability, flexibility, and cost efficiency, while providing users with faster and more reliable service experiences. In line with this trend, the study seeks to investigate the drivers of organizational behavior in switching to cloud-native systems. This study derives relevant factors from prior research and constructs a conceptual framework grounded in the PPM theory. After collecting 292 valid responses, empirical analysis was performed via AMOS 29.0. The results reveal that, with the exception of increased error potential, all variables were found to significantly affect the decision to migrate to cloud-native computing environments. Furthermore, consulting service quality was found to moderate the linkage connecting push and pull elements with the intention to switch to cloud-native systems. The results highlight the pivotal importance of consulting expertise in enabling effective migration processes and provide meaningful insights into the underlying drivers of cloud-native switching.

The importance of recommendation algorithms is underscored by the advancement of AI technology and the growing demand for personalized services. However, there is a lack of empirical studies that analyze differences in algorithmic services across markets. The objective of this study is to draw attention to discrepancy in this field and develop a dual-path model to examine the factors that influence satisfaction with recommendations. By analyzing 641 respondents’ data through Partial Least Squares (PLS), it identifies differences in user attitudes towards algorithms across domains. e-commerce firms and content providers can improve personalized recommendations through the research, which highlights the importance of understanding consumer satisfaction and trust in technology to adapt to evolving AI innovations.