The Society of Convergence Knowledge Transactions ISSN:2287-8920(Print) 융복합지식학회논문지

The purpose of this study was to determine the impact of high school students’ Internet over-dependence on depression/anxiety and to verify whether the parallel mediating effects of academic stress and aggression are significant in this relationship. The subjects of the study were 320 high school students from the 16th Children's Supplementary Survey data of the Korea Welfare Panel, and the research method was correlation analysis, regression analysis and mediation effect analysis using PROCESS macro model 4 with SPSS 24.0. As a result of the study, First, depression/anxiety were positively correlated with Internet over-dependence, academic stress, and aggression, and Internet over-dependence was positively correlated with academic stress and aggression, but the correlation between academic stress and aggression was not significant. Second, Internet over-dependence had a significant direct effect on depression/anxiety, and the partial mediating effect of academic stress was significant. Third, the partial mediating effect of aggression was significant in the relationship between Internet over-dependence and depression/anxiety. We proposed a strategy that considers Internet over-dependence, academic stress, and aggression as an intervention for depression/anxiety in high school students.

The purpose of this study is to derive a definition of AI literacy, establish the elements of AI literacy, and present a framework by applying the Delphi methodology based on the opinions of experts related to AI technology and ethics, digital literacy, and Education. The number of experts who participated in the Delphi survey was the same for the first and second rounds, with a total of 10 people. Data collection was conducted through an online survey from December 20 to December 31, 2024 for the first round and from January 6 to January 12, 2025 for the second round. The contents of questions for the first survey were divided into two areas(definition of AI literacy, definitions of the seven sub-elements of AI literacy framework) and used multiple-choice questions, 5-point scale questions, and open-ended questions. The second survey presented the definition of AI literacy and the definition of seven sub-elements based on the analysis of the feedback collected in the first survey, and evaluated the degree of agreement using a 5-point Likert scale, and included open-ended questions to gather opinions for the development of the final framework. Based on the results of the first and second surveys, the final definition of AI literacy was suggested and a framework that included definitions of the proposed seven elements was developed. Finally, the implications and limitations of the study were described and suggestions for future research were made.

This paper proposes a new method for object detection in occlusion using existing multi-object tracking algorithms. Multi-object tracking is a important algorithm for continuously tracking multiple objects based on real-time video. However, in real-world environments, occlusion can occur, disrupting tracking and making it difficult to continuously detect the same object. In older study, have used re-identification techniques to solve this problem, but their high overhead computational real-time processing. In this paper, we suggest a new method using a non-match retention technique. This method retains detected objects that fail IoU matching for a certain period of time and then re-detects them. The proposed method relies on YOLO-based object detection and a SORT algorithm. We discusses Kalman filtering for implementing the SORT algorithm and a non-match retention method using the IoU matching algorithm. Experimental tests include performance tests on the accuracy of multi-object tracking and buffer size. Experimental results show that the proposed method outperforms existing multi-object tracking methods in terms of frames per second (FPS). In particular, we obtained results that enabled continuous tracking in situations where objects temporarily disappeared and then reappeared in occlusion area.

This study investigates how psychological stress caused by the emergency martial law declared in Korea on December 3, 2024 affected public perceptions of tourism. Using the big-data text-mining platform TEXTOM, data were collected from blogs, cafés, and news articles on Naver, Google, and Daum from December 3, 2024 to July 30, 2025 with keywords including “emergency martial law,” “Lee Jae-myung government,” “national image,” “national brand,” and “tourism.” From 41,730 text records, the top 100 core keywords and their TF-IDF values were extracted, and CONCOR analysis identified four clusters: new government, martial law, tourism, and media. Results show that the martial law incident heightened political and social anxiety, intensified negative emotions such as fear and anger, and temporarily suppressed tourism participation and perceptions. Media dissemination amplified these effects, undermining national image and international tourism demand. The findings provide foundational insights for tourism policy and marketing strategies to cope with political crises and highlight the importance of accurate crisis communication and enhanced digital literacy.

This study aimed to identify factors influencing breastfeeding practices among early postpartum mothers hospitalized after childbirth. The breastfeeding behaviors of mothers admitted to an obstetrics and gynecology hospital in a city in Gyeonggi Province were examined and categorized into exclusive breastfeeding, mixed feeding, and formula feeding. A total of 118 questionnaires were collected to analyze the level of breastfeeding practice according to general, obstetric, and breastfeeding-related characteristics. Correlation analysis identified breastfeeding attitude (r = .466, p < .001), breastfeeding plan (χ² = 21.973, p < .001), and timing of breastfeeding decision (χ² = 9.746, p = .008) as variables significantly associated with breastfeeding practice. These variables were then entered as independent factors in a binary logistic regression analysis. The results indicated that having a breastfeeding plan (OR = 7.244, p = .008) and a positive breastfeeding attitude (OR = 1.073, p = .021) were significant predictors of breastfeeding practice. These findings suggest that establishing a breastfeeding plan and promoting positive attitudes toward breastfeeding may enhance breastfeeding behaviors among postpartum mothers.

Purpose: This study aims to examine the mediating role of college life adaptation in the relationship between intolerance of uncertainty and career decision-making self-efficacy among nursing students. Methods: A descriptive cross-sectional survey was conducted from May 19 to 30, 2025, involving 184 nursing students. The following statistical analyses were performed: t-test, ANOVA, Scheffé test, Pearson’s correlation coefficient analysis, and Hayes’ PROCESS Macro Model 4 (to test the mediating effect). Results: Intolerance of Uncertaainty showed a significant negative correlation with college life adapation, career decision-making self-efficacy. The mediating effect of college life adaptation was found in the effect of intolerance of uncertainty on their carrer decision-making self-efficacy. Conclusion: Through this study, it was confirmed that college life adaptation is a significant factor influencing the career decision-making self-efficacy of nursing students. Therefore, the results of this study can serve as fundamental data for developing strategies to enhance nursing students’ career decision-making self-efficacy.

This study investigated the effects of greenhouse microclimatic conditions on the growth and yield of Perilla (Perilla frutescens) using three years of hydroponic smart farm production data. Pearson correlation and principal component analysis (PCA) were employed to identify the key environmental factors influencing growth and yield. The results showed that dominant factors varied across years, highlighting the complexity of perilla growth models and the need for customized environmental management. Vapor pressure deficit (VPD) consistently had a strong impact: daytime VPD induced water stress, whereas nighttime VPD promoted optimal growth. Year-to-year variability was also observed, with the day-night temperature difference (DIF) being the main factor in the first year, CO2 concentration in the second year, and humidity in the third year. The study emphasizes the importance of nighttime environmental management, confirming that proper control of nighttime humidity and VPD is critical for yield enhancement. All regression models exhibited high explanatory power (R2≥ 89.1%), with nighttime VPD proving to be a reliable indicator for yield prediction. These findings underscore the importance of real-time monitoring and flexible environmental management strategies.

Ensuring reproducibility of model experiments is essential for the practical deployment of AI-based anomaly detection in manufacturing processes; however, existing tools often require manual effort to record and align experiment evidence. This study proposes an experiment management framework for manufacturing anomaly detection that natively integrates MLflow-based experiment tracking into a no-code workflow engine. The framework enables automatic logging of hyperparameters, performance metrics, model artifacts and environment dependency information via GUI-based component composition. We evaluate the framework under an operational reproducibility setting using experiment-tracking completeness and evidence-confirmation cost metrics, demonstrating improved reproducibility support compared with a code-based approach and a general-purpose no-code platform. The proposed design provides reproducibility evidence as a system default, facilitating trustworthy AI operations in manufacturing.

This study addresses the voltage limitation (~1.5 V) of conventional aqueous redox-flow batteries caused by water electrolysis, which restricts the development of high-voltage and high-energy-density ESS systems. To overcome this limitation, non-aqueous electrolytes and new supporting electrolytes are required. The purpose of this work is to synthesize imidazolium-based ionic liquids and evaluate their electrochemical behavior in a non-aqueous V(acac)₃ redox-flow battery. 1,3-dibutylimidazolium bromide (DIBr), 1,3-dibutylimidazolium tetrafluoroborate (DIBF₄), and 1,3-dibutylimidazolium hexafluorophosphate (DIPF₆) were synthesized and evaluated as supporting electrolytes in a non-aqueous vanadium(III) acetylacetonate (V(acac)₃) redox flow battery (RFB). Cyclic voltammetry (CV) indicated that two of the four ionic liquids were electrochemically stable in the V(acac)₃/acetonitrile electrolyte. The stability window was -2.5 to 1.0 V (vs Ag QRE). The concentrations of V(acac)₃ and the supporting electrolytes were both 0.1 M in all electrochemical tests. Under our test conditions, the apparent reversibility trended DIPF₆ > DIBF₄ > tetrabutylammonium hexafluorophosphate (TBAPF₆) > DIBr. These results suggest that the synthesized ionic liquids can serve as potential supporting electrolytes for non-aqueous RFBs. These properties enable higher operating voltages and broaden the usable temperature range, offering advantages for the design of high-energy-density non-aqueous RFB systems. From CV, the estimated cell potentials were approximately 2.2–2.5 V (DIPF₆ > DIBF₄ > TBAPF₆ ≈ DIBr).

The chemical mechanical polishing/planarization (CMP) with the advancement of integrated circuit (IC) technology, processes are evolving to meet the technical requirements of fine line widths and the need for multi-layer wiring. This study reports the synthesis and systematic evaluation of ceria-based CMP slurries intended for interlayer dielectric (ILD) and shallow trench isolation (STI) processes. Ceria (CeO₂) powders were prepared by calcining a cerium-carbonate precursor at 600-800 °C for 5 h, dispersed with a commercial dispersant (D), and processed via vertical milling to obtain aqueous colloidal slurries. Precursor decomposition was examined by thermogravimetric analysis (TGA). The morphology and crystallography of calcined powders were characterized by colorimetry, SEM/TEM, and XRD to estimate particle and crystallite sizes. The primary particles of the synthesized ceria were 15-20 nm in size, while the secondary particles were approximately 200 nm. Slurry physicochemical stability after milling was quantified via pH, conductivity, particle-size distribution, and zeta potential. Polishing performance was assessed by the material removal rate (MRR), computed from pre- and post-polish film thickness, and by planarity, obtained from the post-polish thickness range; common defect modes (ring, scratch, zone, repeat) served as qualitative criteria. The integrated synthesis-formulation-characterization workflow relates precursor calcination and dispersant-assisted dispersion to colloidal stability and wafer-level outcomes, providing a basis for optimizing CeO₂ slurry formulations under stringent CMP requirements. The HA800 slurry showed an absolute zeta potential of about 40 mV, indicating a stable dispersion. Under this condition, the HDP, nitride, and polysilicon films exhibited removal rate/non-uniformity values of 1018 Å/4.08, 302.3 Å/3.89, and 233.5 Å/2.07, respectively, confirming high removal efficiency, low non-uniformity, and high selectivity, so HA800 (calcined at 800 °C in air) was identified as the most suitable slurry for CMP.

This study jointly analyzes the knowledge structure and abstract-level discourse of 1,017 engineering journal articles and reviews (2005–2024; English) retrieved from the Web of Science Core Collection by integrating keyword-based science mapping with Coh-Metrix discourse indices. Author Keywords and Keywords Plus were merged into a single KW_Merged field, from which a co-word network, a centrality–density strategic diagram, and inclusion-index–based thematic evolution were derived, while sentence-length, cohesion, and lexical-diversity indices were computed for the same set of abstracts. The results reveal a three-way structure in which a robotics hub linked to design, system, model, and vision/recognition, a surgical-robotics axis, and an alignment/localization/outcome axis jointly organize engineering research. The strategic diagram and thematic evolution trace a strengthened lineage from minimally invasive surgery to robotic surgery and show increasingly tight couplings between algorithm/design/model terms and outcome/alignment vocabulary. Coh-Metrix analysis further indicates significant late-period (2018–2024) increases in sentence length and referential cohesion, suggesting that structural reorganization is realized in abstracts as denser information packaging and more explicit reference.

This study systematically analyzes the development of digital heritage content utilizing national heritage in South Korea from the 1990s to 2024. Three analytical criteria were applied: representation type, production technology, and production entity. Thirteen major cases led or supported by public institutions were examined to identify period-specific characteristics.The analysis reveals four stages of development from the 1990s to the 2020s. Representation types expanded from realistic to hybrid representation. Production technologies advanced from 3D techniques through immersive technologies to convergent technologies. Production entities diversified from independent government research institutions to cultural institution collaborations, public-private partnerships, and private independent projects.The purposes of utilization progressively expanded from research and information management to preservation and education, industrial application, and artistic exhibition. Digital heritage now functions as a national asset creating cultural and economic value.

The demand for personalized exercise management has been increasing, highlighting the growing importance of feedback systems based on exercise performance records. However, existing services remain largely focused on quantitative indicators such as sets, repetitions, and load, which limits their intuitive understanding for non-experts and reduces their effectiveness in sustaining user motivation. To address these limitations, this study proposes a visual feedback method that reflects exercise performance data through changes in a 3D character’s body shape. The structure of weight training programs was analyzed to define the character’s body into five anatomical regions, and the muscles activated in each exercise—specifically agonist and synergist muscles—were mapped along with set and repetition information to quantify stimulus levels for each region. Furthermore, a cumulative adaptive model was developed based on established principles of muscle hypertrophy to incorporate delayed growth responses, allowing the character’s physical changes to appear in a manner consistent with actual physiological adaptation timelines. By integrating exercise performance records with muscle development mechanisms, this study presents a novel 3D visual feedback approach that enhances users’ comprehension of their training progress and supports sustained engagement. The proposed method demonstrates its potential as an innovative exercise management service that provides intuitive and physiologically meaningful feedback to a wide range of users.

This study was conducted to analyze the impact of digital health literacy, self-leadership, and psychological capital on major satisfaction among nursing students, aiming to identify factors for improving major satisfaction. The results showed that digital health literacy, self-leadership, and psychological capital all showed significant positive correlations with major satisfaction, with psychological capital being the most influential variable. Digital health literacy was found to enhance learning efficiency and confidence in major studies by strengthening the information seeking, understanding, and utilization skills required in a digital-based healthcare environment. Self-leadership was found to promote academic engagement and major adaptability through self-regulation and goal setting. Psychological capital served as a psychological resource that buffered learning stress and career anxiety, enhancing positive attitudes and sustained engagement in the major. Furthermore, clinical practice satisfaction was found to negatively impact major satisfaction, highlighting the need for improvements in the practice environment and guidance system. The final explanatory power was 47.8%, reaffirming that major satisfaction is not a single factor but a complex concept explained by multidimensional psychological and cognitive factors. These results suggest that strengthening digital health literacy, promoting self-leadership, developing psychological capital, and improving the clinical practice environment are necessary to improve major satisfaction among nursing students.

Fake news tends to spread significantly faster than truthful news by stimulating human emotions; however, existing detection research has been limited by primarily utilizing a small number of basic emotional categories, failing to capture complex nuances. This study aims to overcome this limitation by reflecting the comprehensive and multidimensional emotional characteristics inherent in fake news. We specifically utilized EmoBERTa, fine-tuned on the GoEmotions dataset comprising 28 sophisticated emotional labels, to extract granular 'multidimensional emotional features' from the text of the WELFake benchmark dataset. The statistical validity of these extracted features was rigorously verified using T-tests and Mann-Whitney U tests, which proved that the majority of features showed significant differences between fake and real news groups. Finally, we proposed a Random Forest classification model that takes a combined input of contextual vectors extracted via the BERT-base-uncased model and the pre-validated multidimensional sentiment vectors. The proposed hybrid model demonstrated a performance improvement in fake news detection, achieving a Weighted F1-Score increase of 0.14 percentage points compared to the baseline model using only context vectors. This experiment holds significance beyond basic research; by providing statistical evidence of emotional distinctions, it suggests potential for practical application as a valuable tool in the field of social sciences.

Crowdfunding is an effective sales strategy for startups, enabling them to secure funding while simultaneously securing brand awareness and product credibility through the pre-launch of new products. Through crowdfunding, consumers can communicate brand value and build relationships by establishing communication channels with the brand. A key element in increasing success rates in this process is the Unique Selling Proposition (USP), which requires different strategies and implementations for each product category. This study, drawing on prior research, analyzes the differences in USP strategies and the commonalities of successful crowdfunding campaigns across key categories. To achieve this, we conducted an in-depth case study of a food tech company operating in the digital customer experience sector, examining the process from new product selection to market and consumer research and sensory evaluation. The results revealed the importance of a USP that significantly delivers value to consumers in the food industry, and confirmed that this element can enhance sustainability throughout customer engagement and the purchase funnel. We also propose theoretical and practical implications for startups.

The prediction results indicated that investment returns for both DC and IRP schemes are expected to show a consistent pattern: a decline in the first and third quarters and a rise in the second and fourth quarters, remaining within a certain predictable range. Furthermore, when analyzing the overall IRP investment returns, it was observed that the amplitude of fluctuations (both upward and downward) was smaller compared to the returns from non-guaranteed principal products alone. An examination of the asset allocation ratios for DC and IRP revealed that IRP, which provides an annual tax benefit of 3 million KRW more than DC, exhibited a higher allocation ratio. However, despite this higher participation, IRP's actual returns were found to be similar to or even lower than those of DC schemes. This suggests that simply increasing investment participation does not necessarily translate into higher returns, a phenomenon likely attributable to behavioral economic biases such as improper investment decisions due to insufficient knowledge, herd behavior, confirmation bias, and frequent trading.Therefore, based on these findings, it is strongly suggested that concrete pension and investment education provided by both the government and professional investors is essential for the proper and effective management of private pensions.

As climate change increases the uncertainty of natural disasters, traditional structural measures such as dams and levees alone have reached their limitations in fully controlling disaster risks. In response, the international community has emphasized strengthening preventive non-structural measures through the UNDRR Sendai Framework. This study analyzes the current status of non-structural measures in Korea and identifies key limitations, including fragmented implementation and a lack of incentive mechanisms. To address these issues, this research benchmarks the U.S. FEMA Community Rating System (CRS) and proposes a “Korean-style Integrated Non-Structural Disaster Reduction Model” that links quantitative evaluation of disaster-mitigation activities with incentive measures such as insurance premium discounts. This model is expected to encourage voluntary participation in disaster-risk reduction by local governments and residents, thereby contributing to the enhancement of disaster resilience.