• The Korean Journal of Applied Statistics
• /
• v.36 no.6
• /
• pp.515-528
• /
• 2023

With the growth of the e-commerce market, consumers increasingly rely on user reviews to make purchasing decisions. Consequently, researchers are actively conducting studies to effectively analyze these reviews. Among the various methods of sentiment analysis, the aspect-based sentiment analysis approach, which examines user reviews from multiple angles rather than solely relying on simple positive or negative sentiments, is gaining widespread attention. Among the various methodologies for aspect-based sentiment analysis, there is an analysis method using a transformer-based model, which is the latest natural language processing technology. In this paper, we conduct an aspect-based sentiment analysis on multilingual user reviews using two real datasets from the latest natural language processing technology model. Specifically, we use restaurant data from the SemEval 2016 public dataset and multilingual user review data from the cosmetic domain. We compare the performance of transformer-based models for aspect-based sentiment analysis and apply various methodologies to improve their performance. Models using multilingual data are expected to be highly useful in that they can analyze multiple languages in one model without building separate models for each language.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.26 no.4
• /
• pp.145-160
• /
• 2023

The purpose of this study was to compare and analyze diurnal thermal environments using Unmanned Aerial Vehicles(UAV)-derived physical parameters(NDVI, SVF) and ENVI-met modeling. The research findings revealed significant correlations, with a significance level of 1%, between UAV-derived NDVI, SVF, and thermal environment elements such as S↑, S↓, L↓, L↑, Land Surface Temperature(LST), and Tmrt. In particular, NDVI showed a strong negative correlation with S↑, reaching a minimum of -0.52^** at 12:00, and exhibited a positive correlation of 0.53^** or higher with L↓ at all times. A significant negative correlation of -0.61^** with LST was observed at 13:00, suggesting the high relevance of NDVI to long-wavelength radiation. Regarding SVF, the results showed a strong relationship with long-wave radiative flux, depending on the SVF range. These research findings offer an integrated approach to evaluating thermal comfort and microclimates in urban areas. Furthermore, they can be applied to understand the impact of urban design and landscape characteristics on pedestrian thermal comfort.

• Journal of Industrial Convergence
• /
• v.22 no.2
• /
• pp.33-43
• /
• 2024

This study aims to test the relationship between brand embeddedness and cross-selling intention, as well as how customer satisfaction affects as mediator its relationship. The data set includes 258 consumers using a specific smartphone brand and we gather the data by conducting a survey. We use structural equation modeling to verify hypotheses and do nonparametric bootstrap approach for additional analysis of the mediating role of customer satisfaction. The results of analysis are as follows. First, it has been shown that barnd embeddedness positively affects customer satisfaction and cross-buying intention. Second, customer satisfaction is found to have a positive effect on cross-buying. By combining these results and additional analysis, we confirm that the direct effect of brand embeddedness is more effective for cross-buying intention than the indirect effect of customer satisfaction. The results of this empirical analysis show the academic value that brand embeddedness as social capital can be a antecedent for cross-buying intention. In addition, this study suggests the need for brand practitioners to increase the embeddedness of customers related to the brand and consider the mediating role of customer satisfaction together in order to increase cross-buying.

• Korean Journal of Radiology
• /
• v.22 no.6
• /
• pp.994-1004
• /
• 2021

Objective: To extract pulmonary and cardiovascular metrics from chest CTs of patients with coronavirus disease 2019 (COVID-19) using a fully automated deep learning-based approach and assess their potential to predict patient management. Materials and Methods: All initial chest CTs of patients who tested positive for severe acute respiratory syndrome coronavirus 2 at our emergency department between March 25 and April 25, 2020, were identified (n = 120). Three patient management groups were defined: group 1 (outpatient), group 2 (general ward), and group 3 (intensive care unit [ICU]). Multiple pulmonary and cardiovascular metrics were extracted from the chest CT images using deep learning. Additionally, six laboratory findings indicating inflammation and cellular damage were considered. Differences in CT metrics, laboratory findings, and demographics between the patient management groups were assessed. The potential of these parameters to predict patients' needs for intensive care (yes/no) was analyzed using logistic regression and receiver operating characteristic curves. Internal and external validity were assessed using 109 independent chest CT scans. Results: While demographic parameters alone (sex and age) were not sufficient to predict ICU management status, both CT metrics alone (including both pulmonary and cardiovascular metrics; area under the curve [AUC] = 0.88; 95% confidence interval [CI] = 0.79-0.97) and laboratory findings alone (C-reactive protein, lactate dehydrogenase, white blood cell count, and albumin; AUC = 0.86; 95% CI = 0.77-0.94) were good classifiers. Excellent performance was achieved by a combination of demographic parameters, CT metrics, and laboratory findings (AUC = 0.91; 95% CI = 0.85-0.98). Application of a model that combined both pulmonary CT metrics and demographic parameters on a dataset from another hospital indicated its external validity (AUC = 0.77; 95% CI = 0.66-0.88). Conclusion: Chest CT of patients with COVID-19 contains valuable information that can be accessed using automated image analysis. These metrics are useful for the prediction of patient management.

• Journal of Korean Society of Forest Science
• /
• v.113 no.2
• /
• pp.259-270
• /
• 2024

Since 2013, Korea has allocated significant budgets and manpower nationwide to prevent the spread of pine wilt disease and to reduce damage. As a result, the number of damaged trees decreased from 2.18 million in 2014 to 310,000 in 2021. However, the damage has increased again since then. Despite the overall decrease in the number of damaged trees, the scope of the damage continues to expand every year. Previous studies have develope In order to judge the control performance, a quantitative control performance analysis method to objectively evaluate control performance. This method takes into consideration two factors-quantity change and the change in the damage area, which is an area factor. This approach provides a more comprehensive assessment than the control guidelines that only suggest changes in damage grade based on the volume of damaged trees. The expansion of the damage range is also an important factor in analyzing control performance, but previous studies have not reflected this. Therefore, this study calculates the change in the distance of the pine wilt disease boundary area for Gyeongsangbuk-do, where changes in the damage range can easily be observed from year to year. The study then creates application criteria and coefficients and uses them to improves control performance index calculation formula. As a result, it was possible to calculate a quantitative analysis of the control performance, taking into account the changes in the damage range. When the improved formula was applied to 26 cities, counties, and districts in Gyeongsangbuk-do, it slightly decreased or increased compared to the existing calculation formula. This confirmed that the control performance index can change from a positive value (+), indicating increased damage, to a negative value (-), indicating reduced damage.

• Journal of Oriental Neuropsychiatry
• /
• v.35 no.2
• /
• pp.163-175
• /
• 2024

Objectives: To investigate the relationship between rehabilitation treatment, mental state and resilience of stroke patients undergoing rehabilitation by examining the correlation between The Core Seven Emotions Inventory-Short Form (CSEI-s) and the Korean version of the Connor-Davidson Resilience Scale (K-CD-RISC). Methods: All 104 participants (44 diagnosed with stroke who were receiving rehabilitation and 60 without stroke or psychiatric history) completed the CSEI-s, K-CD-RISC, and Questionnaire for stroke symptoms. All data were analyzed using by Statistical Package for the Social Sciences (SPSS) ver. 27.0. Descriptive statistics, chi-square test, t-test, Mann-Whitney U test, Kruskal-Wallis H test, and Pearson correlation coefficient were used for data analysis. Results: As a result of the CSEI-s, compared to the control group, the stroke group showed significantly lower Joy (喜) scores and significantly higher scores for Depression (憂) and Sorrow (悲). With a morbidity period of 12 months or less, the Thought (思) score was significantly higher. The Fear (恐) score was significantly higher when the rehabilitation was initiation more than 4~8 weeks after than that when the treatment was started immediately. Meanwhile, the K-CD-RISC score was significantly higher when rehabilitation was started immediately. In the stroke group, the K-CD-RISC score was positively correlated with Joy (喜) but negatively correlated with Depression (憂) and Fear (恐). In the control group, K-CD-RSIC showed a positive correlation with Joy (喜) but negative correlations with Depression (憂), Sorrow (悲), and Fear (恐). Conclusions: In addition to early rehabilitation treatment, mental approach through Korean medicine psychotherapy is crucial for enhancing the resilience of stroke patients.

• Journal of the Korean Society for information Management
• /
• v.41 no.2
• /
• pp.97-130
• /
• 2024

The current debate in the academic community is on the criteria for predatory journals. Researchers are perplexed about what constitutes a predatory journal. The purpose of this study is to investigate how South Korean researchers discover and evaluate predatory journals. In order to achieve this, we collected 2,484 statements, comprising posts and comments, from Korean researcher communities, namely the Biological Research Information Center (BRIC), Hibrain.net, Phdkim.net, and the Scholarly Ecosystem Against Fake Publication Environment (SAFE). We divided the data into three primary categories-journals, publishers, and researchers-for the topic analysis. For each statement, we assigned 11 in-depth subtopic tags based on these categories. Six main points of contention emerged from the combinations of these sub-topic tags: (1) researchers' confusion about predatory journals and discussions about research performance; (2)(3) researchers' positive and negative perceptions of predatory journals; (4) researchers' evaluation criteria for journal quality and problems associated with the quality of Korean journals; (5) changes in publishing brought about by the introduction of open access (OA) and associated issues; and (6) discussions on broader issues within the academic ecosystem. By using a qualitative approach to examine how South Korean researchers view predatory journals, this study aims to advance basic knowledge of the discourse around them in the communities of domestic researchers.

• CELLMED
• /
• v.14 no.10
• /
• pp.12.1-12.13
• /
• 2024

This study investigated the psychological effects of the creative music fairy tale 'I am a music creator' education program, which was implemented as the convergence educational program of the Seocho Innovation Education District in 2021 for a total of 8 weeks for elementary school students, on Musical Interest, Self-Efficacy, and Self-Esteem. Before verifying the effects after conducting the education program for a total of 8 weeks, a homogeneity verification of the group composition was conducted through a pre-measurement, and it was confirmed that the composition by gender and grade was homogeneous. The results of this study are as follows. First, the results of the pre-test/post-test by research subject showed that the post-test scores in Musical-Interest and Self-Efficacy were slightly higher than the pre-test scores, but there was no significant difference. When Self-Esteem was considered, There was a significant difference, and the post-test scores showed a higher average than the pre-test scores. Second, the results of the pre-test/post-test by gender of the research subject showed that the post-test scores in Musical-Interest, Self-Efficacy, and Self-Esteem for male students were slightly higher than the pre-test scores, but there was no significant difference. The post-test scores in Musical-Interest and Self-Efficacy for female students were slightly higher than the pre-test scores, but there was no real significant difference. Self-Esteem showed a significant difference, and the post-test scores showed a higher average than the pre-test scores. Third, the results of the pre-test/post-test by grade of the research subject showed that the post-test scores in Musical-Interest, Self-Efficacy, and Self-Esteem for second graders were slightly higher than the pretest scores, but there was no real significant difference. In the 3rd grade, there was a significant difference in the sub-factors of Musical-Interest, interest in music classes, and Self-Esteem, and the post-score showed a higher average than the pre-score. In the 4th grade, the post-scores were slightly higher than the pre-scores in Musical-Interest, Self-Efficacy, and Self-Esteem, but there was no significant difference. Through these results, it was confirmed that the integrated music activity approach has a positive influence on elementary school students' Musical-Interest, Self-Efficacy, and Self-Esteem, which are formed in the process of growth and development.

• Korean Journal of Agricultural Science
• /
• v.51 no.2
• /
• pp.187-192
• /
• 2024

This research aimed to investigate the relationship between meat quality and muscle-to-fat ratio in specific cuts of pork (pork belly and Boston butt) utilizing magnetic resonance imaging (MRI). Twenty-eight pigs were selected, and 24 hours post-slaughter, pork belly and Boston butt samples were individually extracted from the left half carcass for MRI assessment. The MRI scans were reviewed using the Picture Archiving and Communications System. Muscle and fat volumes in the pork belly and Boston butt from the cross-sectional images captured by MRI were estimated using Vitrea workstation version 7. Subsequently, these data were processed using Vitrea post-processing software to automatically determine the volumes, measured in milliliters (mL). Additionally, a three-dimensional reconstruction of the organ being studied was generated. The relationship between regions (pork belly and Boston butt) was assessed using Pearson's correlation coefficient, and statistical analysis was conducted using Graph Pad Prism 8. The muscle-to-fat ratio determined by MRI for pork belly was 1 : 0.64, whereas for Boston butt it was 1 : 0.35. Results of comparing the muscle-fat ratio, the correlation coefficient between pork belly and Boston butt was found to be 0.6127 (R² = 0.3754, p < 0.001) based on MRI analysis. As a result of measuring the muscle-to-fat ratio using MRI as a non-destructive approach, there was a positive correlation between the muscle-to-fat ratios of pork belly and Boston butt.

• Journal of Internet Computing and Services
• /
• v.25 no.4
• /
• pp.109-119
• /
• 2024

This study aims to contribute to the initial research on on-device AI in an environment where generative AI-based services on mobile and other on-device platforms are increasing. To derive success strategies for generative AI-based chatbot services in a mobile environment, over 200,000 actual user experience review data collected from the Google Play Store were analyzed using the LDA topic modeling technique. Interpreting the derived topics based on the Information System Success Model (ISSM), the topics such as tutoring, limitation of response, and hallucination and outdated informaiton were linked to information quality; multimodal service, quality of response, and issues of device interoperability were linked to system quality; inter-device compatibility, utility of the service, quality of premium services, and challenges in account were linked to service quality; and finally, creative collaboration was linked to net benefits. Humanization of generative AI emerged as a new experience factor not explained by the existing model. By explaining specific positive and negative experience dimensions from the user's perspective based on theory, this study suggests directions for future related research and provides strategic insights for companies to improve and supplement their services for successful business operations.