• Nuclear Engineering and Technology
• /
• v.56 no.4
• /
• pp.1501-1512
• /
• 2024

Plasma disruption in tokamak experiments is a challenging issue that causes damage to the device. Reliable prediction methods are needed, but the lack of full understanding of plasma disruption limits the effectiveness of physics-driven methods. Data-driven methods based on supervised learning are commonly used, and they rely on labelled training data. However, manual labelling of disruption precursors is a time-consuming and challenging task, as some precursors are difficult to accurately identify. The mainstream labelling methods assume that the precursor onset occurs at a fixed time before disruption, which leads to mislabeled samples and suboptimal prediction performance. In this paper, we present disruption prediction methods based on anomaly detection to address these issues, demonstrating good prediction performance on J-TEXT and EAST. By evaluating precursor onset times using different anomaly detection algorithms, it is found that labelling methods can be improved since the onset times of different shots are not necessarily the same. The study optimizes precursor labelling using the onset times inferred by the anomaly detection predictor and test the optimized labels on supervised learning disruption predictors. The results on J-TEXT and EAST show that the models trained on the optimized labels outperform those trained on fixed onset time labels.

• Information Systems Review
• /
• v.20 no.4
• /
• pp.1-23
• /
• 2018

A smart phone has been widely spread around world and makes people enjoy online shopping in any time and any place. Recently it also changes the distribution environment. O2O (Online-to-Offline) service becomes new normal due to its convenience of ease shopping of product and services. O2O service market shows steady and steep growth, It is reported that, however, 80% of the businesses has been discontinued within the first year because of unstable business models, customer dissatisfaction and distrust of service. Therefore, it is very important research issue to find out influential factors promoting continuous usage intention of O2O service. Previous study shows that it only considers online characteristics and lack of analysis about offline characteristics and social impact factors. The purpose of this paper is to find out continuous usage intention factors of O2O services by literature review, case analysis, and empirical test. A comprehensive research model and related hypothesis are developed and tested by using a structural equation, Survey was carried out among users who have used O2O service including payment service for at least once. Finally 611 samples are selected out of total 813 surveys. The result shows that the model is theoretically proved and 12 out of 17 hypotheses are accepted. The contribution of this paper is that it provides a new theoretical research model about continuous usage intention factors as well as practical guidelines about promoting continuous usage and growth strategies of O2O service.

• The Journal of Engineering Geology
• /
• v.34 no.2
• /
• pp.229-248
• /
• 2024

This investigated the hydrogeochemical and isotopic characteristics of geothermal waters, groundwaters, and surface waters in Dongrae-gu, Busan, South Korea, in order to determine the origins of the salinity components in the geothermal waters, and their formation mechanisms and heat sources The geothermal waters are Na-Cl-type, distinct from surrounding groundwaters (Na-HCO₃^- and, Ca-HCO₃^- (SO₄, Cl)-type) and surface waters (Ca-HCO₃(SO₄, Cl)-type). This indicates the geothermal waters formed at depth as compared with the groundwaters. δ¹⁸O and δD values of the geothermal waters are relatively depleted as compared with the groundwaters, due to altitude effects and deep circulation of the geothermal waters. Helium and neon isotope ratios (³ He/⁴He and, ⁴He/²⁰Ne) of the geothermal waters plot on a single mixing line between mantle (³He = 3.76~4.01%) and crust (⁴He = 95.99~96.24 %), indirectly suggesting that the heat source is due to the decay of radioactive elements in rocks. The geothermal reservoir temperatures were calculated using the silica-enthalpy and Giggenbach models, yielding values of 82~130℃, and the depth of the geothermal reservoir is estimated to be 1.7~2.9 km below the surface. The correlation between Cl/Na and Cl/HCO₃ for the Dongrae geothermal waters requires the input of salty water. The supply of saline composition is interpreted due to the dissolution of residual paleo-seawater.

• Journal of Korea Water Resources Association
• /
• v.57 no.6
• /
• pp.421-435
• /
• 2024

In April 2023, a wildfire broke out in Gangneung located in the east coast region due to the influence of the Yanggang-local wind. In this study, GIS-based RUSLE(Revised Universal Soil Loss Equation) and SEMMA (Soil Erosion Model for Mountain Areas) were used to evaluate the erosion rate due to vegetation recovery in a small watershed of the Gangneung WUI(Wildland-Urban Interface) fire. The small watershed of WUI fire has a low altitude range of 10-30 m and the average slope of 10.0±7.4° which corresponds to a gentle slope. The soil texture was loamy sand with a high organic content and the deep soil depth. As herbaceous layer regenerated profusely in the gully after the wildfire, the NDVI (Normalized Difference Vegetation Index) reached a maximum of 0.55. Simulation results of erosion rates showed that RUSLE ranged from 0.07-94.9 t/ha/storm and SEMMA ranged from 0.24-83.6 t/ha/storm. RUSLE overestimated the average erosion rate by 1.19-1.48 times compared to SEMMA. The erosion rates were estimated to be high in the middle slope where burned pine trees were widely distributed and the slope was steep and to be relatively low in the hollow below the gully where herbaceous layer recovers rapidly. SEMMA showed a rapid increase in erosion sensitivity under at certain vegetation covers with NDVI below 0.25 (Ic = 0.35) on post-fire hillslopes. Gentle slopes with high organic content and rapid recovery of natural vegetation had relatively low erosion rate compared to steep slopes. As subsequent infrastructure and human damages due to sediment disaster by heavy rain is anticipated in WUI fire areas, the research results may be used as basic data for targeted management and decision making on the implementation of emergency treatment after the wildfire.

• Journal of Nutrition and Health
• /
• v.57 no.1
• /
• pp.53-64
• /
• 2024

Purpose: Mitochondria play a crucial role in preserving skeletal muscle mass, and damage to mitochondria leads to muscle mass loss. This study investigated the effects of oxypeucedanin hydrate, a furanocoumarin isolated from Angelica dahurica radix, on myogenesis and mitochondrial function in vitro and in zebrafish models. Methods: C2C12 myotubes cultured in media containing 0.1, 1, 10, or 100 ng/mL oxypeucedanin hydrate were immunostained with myosin heavy chain (MHC), and then multinucleated MHC-positive cells were counted. The expressions of markers related to muscle differentiation, muscle protein degradation, and mitochondrial function were determined by quantitative reverse transcription polymerase chain reaction. To investigate the effects of oxypeucedanin hydrate on mitochondrial dysfunction, Tg(Xla.Eef1a1:mito-EGFP) zebrafish embryos were treated with 5-fluorouracil, leucovorin, and irinotecan (FOLFIRI) with or without oxypeucedanin hydrate and analyzed for mito-EGFP intensity and mitochondrial length. Results: Oxypeucedanin hydrate significantly increased MHC-positive multinucleated myotubes (≥ 3 nuclei) and increased the expression of the myogenic marker myosin heavy chain 4. However, it decreased the expressions of muscle-specific RING finger protein 1 and muscle atrophy f-box (markers of muscle protein degradation). Furthermore, oxypeucedanin hydrate enhanced the expressions of markers of mitochondrial biogenesis (peroxisome proliferator-activated receptor-gamma coactivator 1 alpha, transcription factor a mitochondrial, succinate dehydrogenase complex flavoprotein subunit A, and cytochrome c oxidase subunit 1) and mitochondrial fusion (optic atrophy 1). However, it reduced the expression of dynamin-related protein 1 (a mitochondrial fission regulator). Consistently, oxypeucedanin hydrate reduced FOLFIRI-induced mitochondrial dysfunction in the skeletal muscles of zebrafish embryos. Conclusion: The study indicates that oxypeucedanin hydrate promotes myogenesis by improving mitochondrial function, and thus, suggests oxypeucedanin hydrate has potential use as a nutritional supplement that improves muscle mass and function.

• Journal of Korean Society of Forest Science
• /
• v.113 no.1
• /
• pp.97-106
• /
• 2024

Evaluating carbon (C) and nitrogen (N) inputs from litterfall is important for soil nutrient management to enhance forest productivity and to understand the mechanisms of nutrient cycling in forest ecosystems. This study was conducted to compare C and N inputs from litterfall components of Cryptomeria japonica D. Don an d Chamaecyparis obtusa Endlicher planted in adjacent sites in the Jinju Research and Experimental Forests in Gyeongsangnam-do, South Korea. Litterfall into litter traps was collected at three-month intervals between December 2020 and December 2021, and the C and N concentrations of the litterfall components were measured. Litterfall amounts were not significantly different between the plantations, except for reproductive litterfall components. Litterfall accumulation peaked between December and March. The litterfall C concentration in the needle and seed litterfall was significantly higher for C. obtusa than for C. japonica. By contrast, the C concentrations in needle and flower litterfall differed seasonally. The mean N concentration of needle litterfall was significantly higher in C. japonica (0.96%) and C. obtusa collected between June and September (1.01%) than in the other seasons (C. japonica: 0.43%; C. obtusa: 0.53%). Carbon and N inputs in both plantations were highest in needle litterfall collected from December to March and lowest in needle litterfall collected from June to September. Annual C input by litterfall was similar between the plantations (C. japonica: 3,054 kg C ha^-1 yr^-1; C. obtusa: 3,129 kg C ha^-1 yr^-1), whereas total N input was higher for C. japonica (46.93 kg N ha^-1 yr^-1) than for C. obtusa (25.17 kg N ha^-1 yr^-1). The higher N input in the C. japonica plantation than in the C. obtusa plantation was associated with the input of reproductive components. These results could be applied to improve stand-scale models of C and N cycling by litterfall components in C. japonica an d C. obtusa plantations.

• Journal of Korea Water Resources Association
• /
• v.57 no.3
• /
• pp.181-193
• /
• 2024

The increasing atmospheric imbalance caused by climate change leads to an elevation in precipitation, resulting in a heightened frequency of flooding. Consequently, there is a growing need for technology to detect and monitor these occurrences, especially as the frequency of flooding events rises. To minimize flood damage, continuous monitoring is essential, and flood areas can be detected by the Synthetic Aperture Radar (SAR) imagery, which is not affected by climate conditions. The observed data undergoes a preprocessing step, utilizing a median filter to reduce noise. Classification techniques were employed to classify water bodies and non-water bodies, with the aim of evaluating the effectiveness of each method in flood detection. In this study, the Otsu method and Support Vector Machine (SVM) technique were utilized for the classification of water bodies and non-water bodies. The overall performance of the models was assessed using a Confusion Matrix. The suitability of flood detection was evaluated by comparing the Otsu method, an optimal threshold-based classifier, with SVM, a machine learning technique that minimizes misclassifications through training. The Otsu method demonstrated suitability in delineating boundaries between water and non-water bodies but exhibited a higher rate of misclassifications due to the influence of mixed substances. Conversely, the use of SVM resulted in a lower false positive rate and proved less sensitive to mixed substances. Consequently, SVM exhibited higher accuracy under conditions excluding flooding. While the Otsu method showed slightly higher accuracy in flood conditions compared to SVM, the difference in accuracy was less than 5% (Otsu: 0.93, SVM: 0.90). However, in pre-flooding and post-flooding conditions, the accuracy difference was more than 15%, indicating that SVM is more suitable for water body and flood detection (Otsu: 0.77, SVM: 0.92). Based on the findings of this study, it is anticipated that more accurate detection of water bodies and floods could contribute to minimizing flood-related damages and losses.

• Journal of Venture Innovation
• /
• v.7 no.1
• /
• pp.1-15
• /
• 2024

This study aims to elucidate the characteristics of big tech platform companies' sports content business in an environment of rapid digital transformation. Specifically, this study examines the market structure of big tech platform companies with a focus on Amazon, revealing the role of sports content within this structure through an analysis of Amazon's sports marketing business and provides an outlook on the sports content business of big tech platform companies. Based on two-sided market platform business models, big tech platform companies incorporate sports content as a strategy to enhance the value of their platforms. Therefore, sports content is used as a tool to enhance the value of their platforms and to consolidate their monopoly position by maximizing profits by increasing the synergy of platform ecosystems such as infrastructure. Amazon acquires popular live sports broadcasting rights on a continental or national basis and supplies them to its platforms, which not only increases the number of new customers and purchasing effects, but also provides IT solution services to sports organizations and teams while planning and supplying various promotional contents, thus creates synergy across Amazon's platforms including its advertising business. Amazon also expands its business opportunities and increases its overall value by supplying live sports contents to Amazon Prime Video and Amazon Prime, providing technical services to various stakeholders through Amazon Web Services, and offering Amazon Marketing Cloud services for analyzing and predicting advertisers' advertising and marketing performance. This gives rise to a new paradigm in the sports marketing business in the digital era, stemming from the difference in market structure between big tech companies based on two-sided market platforms and legacy global companies based on one-sided markets. The core of this new model is a business through the development of various contents based on live sports streaming rights, and sports content marketing will become a major field of sports marketing along with traditional broadcasting rights and sponsorship. Big tech platform global companies such as Amazon, Apple, and Google have the potential to become new global sports marketing companies, and the current sports marketing and advertising companies, as well as teams and leagues, are facing both crises and opportunities.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.19 no.3
• /
• pp.53-68
• /
• 2024

Generative AI is receiving a lot of attention around the world, and ways to effectively utilize it in the business environment are being explored. In particular, since the public release of the ChatGPT service, which applies the GPT-3.5 model, a large language model developed by OpenAI, it has attracted more attention and has had a significant impact on the entire industry. This study focuses on the emergence of Generative AI, especially ChatGPT, which applies OpenAI's GPT-3.5 model, to investigate its impact on the startup industry and compare the changes that occurred before and after its emergence. This study aims to shed light on the actual application and impact of generative AI in the business environment by examining in detail how generative AI is being used in the startup industry and analyzing the impact of ChatGPT's emergence on the industry. To this end, we collected company information of generative AI-related startups that appeared before and after the ChatGPT announcement and analyzed changes in industry, business content, and investment information. Through keyword analysis, topic modeling, and network analysis, we identified trends in the startup industry and how the introduction of generative AI has revolutionized the startup industry. As a result of the study, we found that the number of startups related to Generative AI has increased since the emergence of ChatGPT, and in particular, the total and average amount of funding for Generative AI-related startups has increased significantly. We also found that various industries are attempting to apply Generative AI technology, and the development of services and products such as enterprise applications and SaaS using Generative AI has been actively promoted, influencing the emergence of new business models. The findings of this study confirm the impact of Generative AI on the startup industry and contribute to our understanding of how the emergence of this innovative new technology can change the business ecosystem.

• Journal of the Korean earth science society
• /
• v.45 no.3
• /
• pp.239-259
• /
• 2024

In this study, a qualitative meta-analysis was conducted on research papers on earth science education to derive knowledge of students' understanding of specific science topics-greenhouse effect, global warming, and climate change-within the context of collective Pedagogical Content Knowledge (PCK). Twenty-two research papers addressing students' alternative conceptions (misconceptions) about these topics were selected and analyzed for their respective definitions, causes (mechanisms), and impacts. Semantic network analysis and a mental model framework were applied to synthesize the findings. The meta-analysis revealed several key insights: (1) Regarding the greenhouse effect, students often used the terms "greenhouse effect" and "global warming" interchangeably, lacked knowledge about the types of greenhouse gases, and misunderstood their roles. They commonly associated the greenhouse effect with environmental pollution or changes in the ozone layer, failing to recognize its relation to the heat balance between the surface and atmosphere. (2) Concerning global warming, students confused it with sea level rise and linked it to pollution, ozone layer changes, and glacier melting. They understood global warming as a disruption of the heat balance between the surface and atmosphere but had misconceptions about its environmental impacts. (3) In terms of climate change, students used the term interchangeably with global warming, weather change, and climate anomalies. They associated climate change with atmospheric pollution and ozone layer depletion but misunderstood its environmental impacts. As result, three mental models-categorical, mechanistic, and hierarchical misconceptions-were identified as collective PCK. The implications for enhancing earth science teachers' PCK were discussed based on these findings.