Earticle

The rapid expansion of artificial intelligence (AI) into education and industry demands that learners not only acquire technical skills but also develop the ability to critically evaluate AI outputs and navigate ethical considerations. This study constructs a Generative AI (GAI) competency framework for university students and empirically validates its structure. We designed a 35-item survey instrument encompassing seven subscales— Cognitive Understanding (CU), Technical Skill (TS), Critical Evaluation (CE), Attitudinal Openness (AO), Self-Efficacy & Adaptability (SE), Collaborative Ability (CA), and Ethical Responsibility (ER)—and collected responses from 387 Korean undergraduate students. Exploratory factor analysis (EFA) affirmed a seven-factor structure, with high internal consistency across all subscales (Cronbach’s α = 0.842–0.876). By standardizing subscale scores, we developed a roadmap index with percentile and T-score norms. Quartile comparisons revealed that students in the top quartile demonstrated significantly higher CE and AO scores than those in the bottom quartile (Cohen’s d > 1). We provide guidelines for applying this index in curriculum design, advising, and policy development to foster responsible and effective use of GAI tools relationships [1–4].

We aim to develop and provide initial psychometric validation of a multidimensional Career Competency Scale (CCS) integrating personality, interests, work values, core competencies, and AI-adaptability, and to demonstrate its practical utility for generating data-driven career roadmaps for college students. Rapid advances in artificial intelligence and automation have reshaped the skills that university students need for future careers. Traditional career counselling tools often focus on interests or personality alone, overlooking trainable competencies such as analytical thinking, problem solving, communication, collaboration, digital literacy, and adaptability. We therefore developed a five-domain scale—personality (10 items), interests (30), work values (18), twenty-first-century core competencies (36), and career adaptability/AI competencies (35)—totaling 129 items. The instrument was administered online to 387 Korean undergraduates, and its reliability, factor structure, correlation patterns, and cluster profiles were examined Exploratory factor analysis extracted 36 sub‑factors; the scree plot showed an elbow around five factors, corresponding to the conceptual domains relationships [8]. Cronbach’s α ranged from 0.43 (TIPI) to 0.86 (RIASEC) across sections, while subscale alphas ranged from 0.79 to 0.91 relationships [1]. K-means clustering revealed three profiles—balanced, interpersonal–adaptive, and technical–logical. These findings provide initial validity evidence for the multidimensional scale and offer an empirical basis for designing personalised career roadmaps in higher-education settings.

This study empirically examined the effect of visual similarity in Private Brand (PB) product packaging on consumers’ cognitive dissonance and purchase intention. Based on the theoretical framework of selective perception theory and regulatory focus theory, the research experimentally analyzed the cognitive and emotional responses influencing consumer attitude formation. A total of 111 participants were exposed to PB and National Brand (NB) products with manipulated levels of visual similarity, and the effects on cognitive dissonance, brand trust, and purchase intention were measured. The results showed that participants perceived PB packages as visually similar to NB ones; however, such similarity induced negative emotions and reduced purchase intention. In particular, consumers with a prevention-focused orientation experienced stronger cognitive dissonance. These findings suggest that while visual similarity may temporarily enhance brand recognition, it can undermine consumer trust and satisfaction in the long term. Therefore, PB package design should pursue differentiated visual strategies and strengthen brand identity rather than simple imitation to enhance consumer satisfaction. This study provides fundamental insights for establishing future PB premium strategies and package design directions.

The dog organic feed market is rapidly growing due to various factors such as humanization phenomenon, eco-friendly/sustainability trend, and increased demand for customized pet food. For the purpose of research, the purpose of this study is to demonstrate the positive (+) effect of dog organic feed brands on safety and intention to pay and the mediating effect of safety. In order to demonstrate the positive effect of dog organic feed brands on safety and intention to pay, and the mediating effect of safety, an online survey and multiple regression mediating effect were conducted on 175 dog organic feed users in the Seoul metropolitan area for about three weeks from June 21, 2025. As a result of the analysis, it was found that the premium image had a strong positive (+) effect on the safety perception of dog food (β=0.343), and that consumers use the high-end image as a key criterion for safety evaluation. This suggests that brands should build a premium and reliable image through an integrated marketing strategy and increase safety awareness by emphasizing eco-friendly and functional raw materials. In addition, 'healthy materials' were identified as the most powerful and significant key factor influencing consumers' willingness to pay (β=0.405). Since consumers make the quality and health effects of raw materials the top priority of their willingness to pay, brands should create synergy through luxurious packaging and storytelling along with emphasizing the quality and efficacy of products. In conclusion, the recognition of 'safety' for dog organic feed brands plays an important role in strengthening the willingness to pay.

As cyber threats continue to evolve in sophistication, there is urgent need for intelligent, adaptive and context aware intrusion detection systems. In this paper, we present an intrusion detection framework that employs deep learning models to detect anomalies in network traffic using custom dataset. The dataset was constructed in a controlled lab environment using various intrusion attack scenarios such as DoS, SSH abuse and VPN exploitation. Deep learning models were then applied to detect the intrusions. The performance of the models in performing detection tasks were evaluated using metrics of accuracy, precision, recall and F1- score. The results that were obtained indicate that hybrid model achieved the best results with overall accuracy of 0.99 followed by transformer (0.98) and LSTM model (0.97) being the last. This study highlights the potential of leveraging well designed custom IDS datasets and deep learning techniques to enhance intrusion detection mechanisms thereby providing a robust framework for intrusion detection applications.

In the digital environment, AI algorithms serve as the core technology for optimizing user experience; however, biased data learning and design flaws often exacerbate the echo chamber phenomenon by overemphasizing specific information. This limits information diversity and contributes to social conflict and polarization. Accordingly, this study aims to analyze the bias inherent in AI algorithms and explore information design strategies to mitigate such bias, thereby establishing an information delivery framework that prevents echo chambers. The research methodology for studying information design to counteract echo chambers caused by AI algorithmic bias consists of three stages. First, a structural analysis is conducted to examine how AI algorithm bias forms and reinforces echo chambers. This includes identifying problems that occur in the filtering and recommendation processes and exploring design-based approaches to minimize bias. Second, from the perspective of information design, strategies for preventing echo chambers are investigated. Visual, structural, and interactive design elements are utilized to maximize information diversity and enable users to access balanced perspectives through intuitive interface designs. Finally, a practical design framework is developed that can be applied to public policy platforms, media systems, and memory-based AI interfaces, ensuring real-world usability. Through this approach, the study seeks to establish a design model that enables AI technologies to regain social trust and contribute to a fair and inclusive information ecosystem. Ultimately, this research emphasizes that information design can play a pivotal role in aligning AI systems with human values, fostering transparency, diversity, and equity in the digital information environment.

The rapid advancement of generative artificial intelligence (AI) has reshaped video advertising by shifting creative production from manual craftsmanship to algorithmic collaboration. Early AI-generated advertisements were limited to sequential image generation with minimal narrative coherence, but recent developments in multimodal systems have enabled cinematic continuity, emotional expression, and character stability. This study examines how advertising expression formats have evolved in this context by comparing 20 traditional and 20 AI-generated video advertisements. A mixed-method analysis—combining qualitative content evaluation and quantitative frequency comparison—was conducted to classify six expressive formats: storytelling, character-based, reversal, metaphorical, exaggerated, and emotional types. The results show that AI-generated advertisements increasingly emphasize storytelling and emotional engagement, while metaphor and exaggeration appear less frequently. These shifts suggest that generative AI strengthens narrative coherence and affective immersion but continues to face limitations in producing abstract or humor-based expressions. Qualitative analysis further identifies three adaptive creative types—Inherited, Transformed, and Fusion—that illustrate how AI reinterprets or hybridizes traditional expressive structures. The findings indicate that generative AI expands rather than replaces human creativity, giving rise to a hybrid model of authorship in which human intention and algorithmic generation operate jointly. This study contributes a conceptual framework for understanding the expressive transformation of video advertising in the AI era and highlights the growing importance of AI literacy, prompt engineering, and ethical considerations for future creative practice.

This study examined the effects of virtual reality (VR) experience on nursing students’ flow, empathy, interactivity, cyber sickness, and learning satisfaction. The study also analyzed correlations among these variables and compared them based on participants' VR-related characteristics. A total of 97 nursing students at a university completed a questionnaire. Data was analyzed using descriptive SPSS 27.0. Learning satisfaction showed significant positive correlations with flow, empathy, and interactivity, while it had a significant negative correlation with cyber sickness. Students who perceived VR/AR education as necessary showed higher levels of flow, empathy, and learning satisfaction. Those who expected VR training to be helpful also reported significantly higher flow, empathy, and learning satisfaction. These findings suggest that VRbased learning positively affects learners' psychological engagement in nursing education and underscore the importance of instructional design to maximize its effectiveness. I believe that in order to effectively utilize VR education as a strategy that enhances learners' cognitive and emotional experiences, we must carefully consider technical inconveniences and learner characteristics during instructional design.

After COVID-19, travel patterns have shifted from group tours to individual tourists who plan their own trips. Therefore, the development of generative AI-based, personalized smart tourism services for individual tourists is necessary. We are developing a smart tourism service platform that provides tourism information, travel planner services, and tour guide services using an AI-based chatbot service. In this paper, we develop the smart tourism QA chatbot service for individual tourists, using ChatGPT-4o-mini and RAG. The RAG system must accurately identify the intent of the user's question and generate high-quality prompts to mitigate the Hallucination and Sovereign AI problems inherent in generative AI. In this paper, we propose the RAG system that uses existing BERT-based NER and DST models, the tourism information MySQL DB, and the tourism information knowledge base implemented with a Neo4J graph DB to improve accuracy. The NER and DST models manage the user's conversation state, identify the intent of the question, and search the tourism information knowledge base to generate high-quality prompts. The performance of the proposed RAG system and ChatGPT-4o-mini is analyzed using the previously developed tourism information QA training dataset, and the results show an excellent performance of 99.54%. The sovereign AI problem can be mitigated by using the NER model that applies forbidden words to identify the user's question intent. The smart tourism information QA service proposed in this paper can be used to develop travel planner and tour guide services.

This study conducts a theoretical analysis of film color generated by Artificial Intelligence through the lens of Johann Wolfgang von Goethe's Zur Farbenlehre. It identifies the dominant digital color paradigm as Newtonian, a framework based on the quantitative manipulation of light as objective data. Conversely, this study posits Generative AI, specifically controllable Diffusion Models, as a technology capable of realizing a Goethean paradigm that is phenomenological, dynamic, and holistic. The research aims to deconstruct the analytic framework of digital color grading and re-evaluate AI's creative potential within Goethe's principles of polarity, intensification, and totality. The methodology employs a hermeneutic comparative analysis of Goethe's 1810 text and the technical principles of GANs and Diffusion Models. The study concludes that Generative AI mirrors Goethe’s dynamic principles by generating color from the interplay of light and dark, enabling the practical application of concepts like the Urphänomen and Steigerung. This approach offers a path toward achieving a totality of color in cinema that is affectively resonant and aligned with the human cognitive demand for completeness, elevating AI from a tool of data manipulation to a medium of phenomenological experience.

The present study examined socio-psychological factors that influence consumers’ technophobia toward artificial intelligence (AI) service robots. A survey of adult consumers was conducted using a professional research panel, and a hierarchical regression analysis identified key predictors. We found that older individuals and women reported higher levels of technophobia. Rational trust in AI robots reduced technophobia, whereas emotional trust increased it, indicating that the two dimensions of trust play distinct roles. Social connectedness heightened technophobia, while subjective happiness lowered it. Involvement with AI robots showed a positive but non-significant relationship, suggesting a more complex pattern than initially expected. Overall, the findings highlight the importance of considering cognitive, emotional, and social factors simultaneously when explaining fear responses to AI robots and underscore the need to distinguish between rational and emotional trust in theorizing about human–robot relationships. We discuss the practical implications of the findings for robot design and deployment, including strategies to enhance rational trust, calibrate emotional expressiveness, and provide reassuring information in public settings.

This study proposes a design methodology for a self-evolving system that can proactively detect failures occurring in educational information systems and automatically implement corrective measures. Because educational information systems exhibit highly regular and cyclical structures—driven by academic schedules and recurrent user activity patterns. To this end, the study designs an anomaly-detection framework that combines Poisson-distribution–based frequency analysis with regular-expression–based pattern grouping. Detected anomalous events are forwarded to an automated deployment process that generates and applies corrective logic, which is then incorporated into active service instances without downtime through suggested deployment structure. This study presents a reliability-enhancement model that shifts failure management in educational information systems from reactive recovery to a proactive and automated paradigm. In experiments involving 79 stability test cases, the proposed system achieved 88.5% precision in anomaly detection and a 94.9% success rate in automated deployment, demonstrating its practical effectiveness in improving system stability. The approach is broadly applicable not only to educational platforms but also to a wide range of public and private information-system environments.

This paper proposes an end-to-end automated framework for designing and managing hands-on learning programs in engineering education. The proposed system is designed as an end-to-end framework that includes automatic identification of practice-oriented courses, analysis of practicum-related documents, integration of instructor feedback, generation of learner-specific practicum modules, LMS-based coordination of interactive practice activities, and linkage with academic information systems. A case-based evaluation shows that the system has positive effects on improving the efficiency of practicum design, providing learners with tailored practice experiences, and reinforcing consistency in practicum operation. By adopting a data-driven approach to practicum design in engineering education, this study contributes to enhancing practicum quality and establishing an automated cycle that connects the design, implementation, evaluation, and continuous improvement of practicum programs. A case-based evaluation involving experts and undergraduate students qualitatively confirms the feasibility, practical effectiveness, and applicability of the proposed framework.

This study examines how brain drain impacts South Korea's cybersecurity amid shrinking R&D budgets and U.S. tariff-induced financial constraints. OLS, Bayesian, and Monte Carlo analyses consistently show that a 1 percentage-point increase in IT/security talent outflow is associated with 59–76 additional server hacking incidents (preferred specification R² = 0.938). R&D budget and AI index effects are inconsistent due to collinearity (r ≈ 0.9). Financial market volatility exacerbates funding constraints, driving talent outflows. With 40–45 % of domestic cybersecurity talent migrating abroad, SMEs—lacking resources to absorb talent shocks—face substantially elevated risks. This paper concludes that IT workforce retention (via R&D budgets) is more critical than AI adoption (AI index) for cybersecurity resilience. Policy priorities should therefore focus on long-term talent retention and shared specialist pooling for SMEs.

This study aimed to investigate the effects of temporomandibular joint (TMJ) mobilization and tongue strengthening exercises on voice in a patient with motor speech disorder following traumatic brain injury (TBI). A single patient received intervention five times per week for a total of 34 sessions, each lasting 30 minutes. The program consisted of Kaltenborn joint mobilization, the Iowa Oral Performance Instrument (IOPI), and proprioceptive neuromuscular facilitation (PNF) techniques to improve TMJ mobility and strengthen tongue muscles. To examine the effects on voice, acoustic analysis was performed to evaluate vocal parameters, and task performance was measured. First, comparison of acoustic parameters before and after the intervention including fundamental frequency, jitter, shimmer, and noise to harmonics ratio showed improvements in some measures related to speech production. Second, diadochokinetic rate was assessed to evaluate articulatory coordination. Third, maximum phonation time was measured to analyze respiratory control, phonatory pattern, and phonation duration. The findings demonstrated that TMJ mobilization and tongue strengthening exercises positively influenced several aspects of speech and voice production. TMJ mobilization and tongue-strengthening exercises had positive effects on vocal and lingual function in a patient with motor speech disorder following traumatic brain injury.

This study clarifies the concept of nurses’ self-management competency by identifying its attributes, antecedents, and consequences, and proposes a conceptual basis applicable to nursing education and clinical practice. Using Walker and Avant’s eight-step concept analysis (2019), we reviewed studies published from 2010 to 2025 across PubMed, CINAHL, RISS, KISS, and DBpia with the keywords “self-management,” “self-care,” “competency,” and “nurse.” Sixty-eight records were screened; twenty-eight met inclusion criteria for analysis. Results: Five defining attributes were derived—(1) self-awareness, (2) self-regulation, (3) emotional control, (4) resilience, and (5) self-development. Antecedents comprised personal factors (e.g., intrinsic motivation, self-efficacy, health literacy) and organizational factors (e.g., leadership support, training, work environment, feedback culture). Consequences included reduced presenteeism and burnout, improved job satisfaction and organizational commitment, increased health-promoting behaviors, and a strengthened patient-safety culture. Self-management competency is a core capability for sustaining professional performance among nurses. Systematic programs at educational and organizational levels are needed to enhance this competency.

We investigated the restorative effects of collagen peptide (CP) and keratin peptide (KP) treatments on hair fibers damaged through repeated 6% H₂O₂ bleaching. To evaluate the independent and combined influences of protein type and thermal activation, we designed a factorial experiment (CP × KP × heat) and analyzed hair thickness and cuticle morphology using a linear mixed-effects model (LMM) with tress-level random effects. We also performed multiplicity-adjusted comparisons and reported standardized effect sizes to ensure statistical robustness. We found that heat activation significantly enhanced the efficacy of both CP and KP, with the combined c4k1 condition yielding the greatest improvements in fiber diameter and cuticle surface integrity. SEM imaging confirmed reductions in cuticle lifting, fragmentation, and surface irregularities, especially under thermal activation. We demonstrate that protein synergy and heat activation are key determinants in restoring chemically damaged hair, providing scientific evidence for the development of advanced functional hair-care treatments.

Traditional grain syrup (Jochung), saccharified by malt, and modern grain syrup (Moolyut), often corn-based with industrial enzymes, coexist in Korea, but their sensory divergence remains unquantified and poses a challenge to objective quality definition. We aimed to objectively define quality criteria by quantifying the inherent taste profiles of these two syrup types and identifying the dominant factor driving their flavor differences. An electronic tongue (E-tongue) was utilized to measure and analyze the taste profiles of 14 commercial syrups, and the resulting data were processed using principal component analysis (PCA). PCA successfully grouped samples into four distinct clusters. The analysis demonstrated that the primary raw material (rice vs. corn) was the dominant factor shaping flavor, overriding the saccharification method. We quantitatively showed that traditional Jochung possesses a distinctive, appealing complexity of high umami and sourness. Conversely, modern corn-based Moolyut exhibits a simpler profile dominated by pronounced saltiness. This study is the first to employ E-tongue technology to objectively quantify these crucial sensory differences, providing critical quantitative evidence to support the establishment of quality standards for Jochung.

This study is to analyze the change in thinking through a new information program to prevent intentional self-harm. The experimental group conducted a new information program for 63 high school students in C area, and the control group selected 52 high school students from April 2 through June 3, 2025. The comparison between the experimental and the control group before and after the information program was analyzed using the paired t-test. In order to verify whether there is a significant difference compared to the average of the pre-test, each pre-test score was controlled as a covariate. As a result of comparing the mean difference between the pre-post experimental and control groups, the experimental group decreased in the risk of intentional self-harm after the information program was implemented(t=2.06, p<.05), while the control group increased(t=-3.52, p<.01). It was found that the experimental group changed positively in intentional self-harm thinking than the control group. To verify the homogeneity of the group, comparing the mean difference between the pre-experiment-control group showed that it was not significant, and it was identified as a homogeneous group. The experimental group showed an overall improvement in post-change than before, while the control group showed little pre-post change. Therefore, it is necessary to develop a family intervention program that reflects socio-cultural characteristics. Furthermore, research should continue to be carried out to enable participation in various levels of target, region and programmes

This study is to identify the factors influencing the quality of oral care by the perception of bad smell in the mouth of pneumonia patients. The subjects of this study conducted a survey of 81 people who visited the internal medicine department of tertiary hospital in the S area. Oral symptoms and function according to the perception of bad smell in the mouth were analyzed by Chi-square test. For the quality of oral care according to the perception of bad smell in the mouth, the t-test was used. Logistic regression analysis was performed on the factors influencing the quality of oral care by the perception of bad smell in the mouth. There were significant differences in chewing food(X2=3.82, p=.014) according to breath awareness, difficulty in pronunciation(X2=9.74, p=.025) and the presence or absence of teeth(X2=5.16, p=.037). There were significant differences in gum bleeding(Х2=3.82, p=.014), toothache(Х2=9.74, p=.025),, tongue or cheeck pain(Х2=5.16, p=.038), and xerostomia(Х2=12.97, p=.029) according to the perception of bad smell. Cox and Snell's explanatory power increased by 24.7%. Nikkei's explanatory power increased by 36.2%. In conclusion, this study will be the basis for the development of oral health programs related to oral health behaviors, dietary habits, toothbrushing habits, and oral health education of pneumonia patients.

This study is to identify the factors that affect the environment of daily life by postoperative inaccuracy in bladder cancer patients. The subjects of the study conducted a survey of 71 people diagnosed with bladder cancer in an outpatient clinic in a urology of tertiary hospital in the C area from May 13 to July 10, 2024. The environment of daily life according to the general characteristics of the subjects was analyzed by t-test and ANOVA. A Scheffè test was used as a post hoc test. Multiple regression analysis was used to understand the influence of general characteristics on the environment of daily life. The results of this study are as follows, Firstly, according to general characteristics, there were significant differences according to monthly income(F=3.92, p=.004), age(F=4.18, p=0.25) in the environment of daily life. Secondly, disease-related characteristics showed significant differences according to metastasis(t=-3.17, p=.042) and recurrence(t=2.96, p=.038). Thirdly, inaccuracy and uncertainty risk assessments were significant positive correlations(r=034, p<.001). Fourthly, the explanatory power of the whole model was 63%. In multiple regression analysis, it was found that monthly income, metastasis, and recurrence significantly explained the environment of daily life. The quality of the environment was significantly explained by uncertainty assessment. Therefore, in order to improve the quality of environment of bladder cancer surgery patients, it is necessary to develop multi-faceted and practical interventions to reduce uncertainty about disease and convert evaluation of uncertainty into positive opportunity evaluation

Regional universities are experiencing a structural crisis driven by demographic decline, rural depopulation, and increasing competition, requiring them to reinvent their public mission. This study aims to develop the Bio-Digital Campus model, which positions the university as an integrated smart hub that supports community health, nutrition, and sustainability. Using a qualitative design, the study employed a PRISMA-guided systematic review of 1,267 records, comparative case analysis of domestic and international smart campus initiatives, and document and policy analysis. A total of 148 studies were synthesized and organized into six thematic domains related to nutrition, health data, robotics, supply chains, energy, and governance. The findings indicate that integrating these systems can enhance metabolic health, local food economies, community engagement, and energy resilience. The model also reveals risks—including engagement decay, data privacy concerns, and supply-chain instability—that necessitate privacy-by-design governance, affordability protections, and inclusive participation. Overall, the Bio-Digital Campus provides a practical blueprint for transforming regional universities into bio-digital wellcare centers that advance the university’s third mission and the UN Sustainable Development Goals.

This study conducts a scoping review to examine how the convergence of blockchain technology and eGovernment can optimize Official Development Assistance (ODA) activities in developing regions by enhancing transparency, accountability, and trust. Guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) protocol, literature searches were conducted across five databases—Google Scholar, Web of Science, Wiley, PubMed, and PLOS—covering publications from 2008 to 2023. Out of 290 records initially identified, only four met the inclusion criteria, indicating a significant research gap in the application of blockchain–eGovernment convergence to ODA and aid management. Findings reveal that while blockchain offers high potential for traceability and transparency in development cooperation, implementation challenges persist, including limited political will, inadequate regulatory frameworks, and a lack of standard guidelines. To address these gaps, this paper proposes an ICT for Transparency (ICT4T) framework, built upon the Hyperledger Fabric architecture, which enables secure, permissioned, and auditable ODA transactions through role-based validation and decentralized consensus. The framework demonstrates how blockchain-enabled eGovernment can strengthen ODA governance, facilitate donor-recipient accountability, and support regional development goals aligned with the Sustainable Development Goals (SDGs). Future studies are recommended to empirically test the proposed model and explore its applicability across different developing regions.

According to recent research results, it has been found that depression treatment helps improve the prognosis of heart disease. This means that depression can negatively affect the recovery process of heart disease patients. In other words, the stress of heart disease patients stimulates sympathetic nerves and increases blood pressure and heart rate, which can worsen heart disease, so appropriate management is essential. In particular, since heart disease and depression are correlated adversely with each other, it is very important to steadily manage both mental and physical health. A recent study result has been published that, if there is depression, the risk of cardiovascular disease may be high, so it is better to receive a heart health check. In this paper, to solve this problem, we performed stress self-diagnosis and HTP psychological tests, proposed algorithms and computer simulations that performed optimal music therapy and art therapy techniques and kidney disease risk prediction for the patient's own stress and constitution conditions.

This study develops a smart, safe modular shelter design integrating AI and IoT technologies, focusing on bus stops as urban data nodes to enhance user safety and convenience through analysis of major domestic and international cases. Domestically, Seoul emphasizes universal design and renewable energy, Incheon digital guidance and accessibility, Daegu CCTV and real-time information, Busan wind-path ventilation and solar power, Gwangju wireless charging and climate control, and Daejeon integrated information with low-floor bus linkage, highlighting common strengths in accessibility, eco-friendliness, and safety. Internationally, Copenhagen features transparent modules, AI detection, and bicycle integration; Singapore AI density prediction and cooling; Amsterdam green roofs and sustainable materials; Dubai climate-adaptive features and Wi-Fi; and Paris multifunctional services with renewable energy, underscoring data-driven operations and environmental sustainability. Based on this analysis, the proposed smart shelter modular design maximizes openness and CPTED effects through transparent materials and curved structures, while automatically controlling temperature, humidity, air quality, lighting, and ventilation via IoT/AI. It integrates wireless charging, smart guidance, greenery, and seating to enhance convenience, realizes sustainability with solar power, LED lighting, and recycled materials, and ensures universal design accessibility via step-free entrances and wheelchair spaces. This design proposal aims to transform smart shelters from mere waiting spaces into integrated smart infrastructure combining safety, environment, and information, thereby driving urban transport innovation and improving citizens' quality of life.

Interest in art appreciation has grown in recent years, yet many individuals—especially those without prior knowledge of art—still find it difficult to interpret and engage with artworks. Feldman’s art critique method, one of the most representative art critique frameworks, provides a structured and systematic art appreciation, but it still remains challenging for non-experts to apply in real-world appreciation contexts. While interactive approaches have been introduced to support art appreciation, most of them still focus on one-directional information delivery, and research on technologies that explore viewers' deeper engagement with art appreciation remains limited. Thus, we propose an AI-based agent that integrates Feldman’s art critique stages into an interactive web-based system for guided art appreciation. To achieve this, a large language model (LLM) integrated with a lightweight RAG pipeline was applied, and custom datasets were created for the RAG. A preliminary study with 10 individuals was conducted to explore user expectations for AI-supported art appreciation. Most participants expressed a preference for interactive systems that provide clear and informative guidance. Based on these insights, the proposed system was designed with the goal of supporting step-by-step interpretation, and contextual understanding of artworks. An exploratory user study with two participants was conducted, revealing positive responses toward the AI-assisted Feldman art critique system.

The proliferation of no-code and low-code automation platforms has fundamentally transformed how organizations approach workflow automation, enabling non-technical users to create sophisticated integrations previously requiring extensive programming expertise. While existing comparative studies predominantly focus on feature inventories and pricing tiers, this paper distinguishes itself through three novel perspectives. First, it employs a strategic deployment sovereignty lens, explicitly examining how cloud-native versus self-hosted architectures align with organizational data governance requirements and infrastructure philosophies—a dimension largely absent in conventional tool comparisons. Second, unlike studies treating AI integration as a peripheral feature, this research positions AI-driven intelligent orchestration as a fundamental architectural paradigm, analyzing how platforms differ in their native AI capabilities versus external service integration approaches. Third, this study provides empirical validation through three realworld use case implementations spanning data synchronization, automated reporting, and AI-augmented customer support, demonstrating practical performance characteristics rather than relying solely on vendor specifications. Through this multi-dimensional framework combining architectural analysis, deployment model implications, AI integration strategies, and scenario-based empirical evidence, the research reveals that platform selection constitutes a strategic decision requiring alignment with organizational infrastructure philosophy, regulatory constraints, technical capability, and AI adoption maturity. This study indicates that Make optimizes for rapid deployment with comprehensive pre-built integrations, n8n provides maximum flexibility and data sovereignty through open-source self-hosting, and Opal demonstrates emerging AI-native capabilities with corresponding maturity trade-offs. These insights enable organizations to move beyond superficial feature comparisons toward strategic tool selection aligned with long-term automation vision and operational requirements.

Large language models have catalyzed significant changes in software development through AI-powered coding assistants. This study examines prompt design strategies and features of three widely-used tools: GitHub Copilot, ChatGPT, and Amazon CodeWhisperer. Through systematic benchmarking using algorithmic problems, API integration tasks, and debugging scenarios, we quantify performance differences across accuracy (71-92%), executability (71-94%), and security vulnerability rates (3-12%). GitHub Copilot demonstrates strongest algorithmic performance (92%), while CodeWhisperer excels in API integration (88%) with lowest security vulnerability rate (3%). We provide evidence-based tool selection guidelines and a practical prompt engineering checklist for developers and educators.

The growing mandate for temperature monitoring in the distribution of temperature-sensitive pharmaceuticals has increased the demand for reliable and user-friendly data logging devices. However, current commercial data loggers face limitations such as high cost, operational errors, physical vulnerability, and inconsistent data accuracy. This study proposes a UX-centered design for an ultra-low-power IoT-based temperature data logger to address these issues. Key design elements—including intuitive interaction, enhanced durability, improved wireless communication, and optimized information delivery—were derived from user requirement analyses and applied to a functional prototype. Additional modeling efforts improved device size, visibility, UI suitability, and mechanical interfaces by refining internal circuit placement and external components such as buttons, charging ports, and displays. Prototype testing in cold-chain conditions demonstrated improved usability, reliability, and operational efficiency. The findings highlight the value of user-centered design in advancing temperature monitoring technologies for pharmaceutical logistics.

This study proposes a conceptual framework for understanding the therapeutic mechanisms and effects of immersive media art through comprehensive literature review of peer-reviewed research published between 2015 and 2024. The introduction establishes the theoretical gap in understanding how immersive media art contributes to therapeutic outcomes beyond traditional art therapy and purely technological interventions. The methodology section describes the integrative literature analysis approach for synthesizing findings across multiple disciplines. Key findings reveal that immersive media art possesses three distinctive therapeutic mechanisms: embodied presence through sensorimotor engagement, interactive narrative development enabling personal meaning-making, and sensory synchronization facilitating emotional regulation. The analysis reveals that these mechanisms operate synergistically to create unique psychological states distinct from traditional therapeutic modalities. The conclusion proposes an integrated theoretical framework combining artistic principles, technological capabilities, and therapeutic objectives, while identifying directions for future empirical research in clinical and non-clinical settings.