• 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.2
• /
• pp.39-62
• /
• 2018

Thanks to the growth of computing power and the recent development of data analytics, researchers have started to work on the data produced by users through the Internet or social media. This study is in line with these recent research trends and attempts to adopt data analytical techniques. We focus on the impact of "internal marketing" factors on firm performance, which is typically studied through survey methodologies. We looked into the job review platform Jobplanet (www.jobplanet.co.kr), which is a website where employees and former employees anonymously review companies and their management. With web crawling processes, we collected over 40K data points and performed morphological analysis to classify employees' reviews for internal marketing data. We then implemented econometric analysis to see the relationship between internal marketing and market capitalization. Contrary to the findings of extant survey studies, internal marketing is positively related to a firm's market capitalization only within a limited area. In most of the areas, the relationships are negative. Particularly, female-friendly environment and human resource development (HRD) are the areas exhibiting positive relations with market capitalization in the manufacturing industry. In the service industry, most of the areas, such as employ welfare and work-life balance, are negatively related with market capitalization. When firm size is small (or the history is short), female-friendly environment positively affect firm performance. On the contrary, when firm size is big (or the history is long), most of the internal marketing factors are either negative or insignificant. We explain the theoretical contributions and managerial implications with these results.

• Annals of Hepato-Biliary-Pancreatic Surgery
• /
• v.27 no.4
• /
• pp.403-414
• /
• 2023

Backgrounds/Aims: Pancreatoduodenectomy (PD) is recommended in fit patients with a carcinoma (PDAC) of the pancreatic head, and a delayed resection may affect survival. This study aimed to correlate the time from staging to PD with long-term survival, and study the impact of preoperative investigations (if any) on the timing of surgery. Methods: Data were extracted from the Recurrence After Whipple's (RAW) study, a multicentre retrospective study of PD outcomes. Only PDAC patients who underwent an upfront resection were included. Patients who received neoadjuvant chemo-/radiotherapy were excluded. Group A (PD within 28 days of most recent preoperative computed tomography [CT]) was compared to group B (> 28 days). Results: A total of 595 patents were included. Compared to group A (median CT-PD time: 12.5 days, interquartile range: 6-21), group B (49 days, 39-64.5) had similar one-year survival (73% vs. 75%, p = 0.6), five-year survival (23% vs. 21%, p = 0.6) and median time-to-death (17 vs. 18 months, p = 0.8). Staging laparoscopy (43 vs. 29.5 days, p = 0.009) and preoperative biliary stenting (39 vs. 20 days, p < 0.001) were associated with a delay to PD, but magnetic resonance imaging (32 vs. 32 days, p = 0.5), positron emission tomography (40 vs. 31 days, p > 0.99) and endoscopic ultrasonography (28 vs. 32 days, p > 0.99) were not. Conclusions: Although a treatment delay may give rise to patient anxiety, our findings would suggest this does not correlate with worse survival. A delay may be necessary to obtain further information and minimize the number of PD patients diagnosed with early disease recurrence.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.69 no.1
• /
• pp.34-48
• /
• 2024

The objective of this study was to evaluate the safety of biodegradable mulching films in soybean (Glycine max) cultivation by measuring their effects on crop growth and yield, film decomposition and soil chemical and physical properties. In 2022 and 2023, plant height, branch number, chlorophyll contents, yield components, and yield of soybean did not vary significantly in areas using PE films and biodegradable mulching films. The light transmission rate of the biodegradable mulching films ranged from 6.4 to 15.8% when measured 112 days after soybean transplanting, and was higher, on average, in 2023 than in 2022. In both years, degradation of the biodegradable mulching films began 20 days after soybean transplantation and increased over time. In addition, remains of biodegradable mulching films were present in fields at soybean harvest and remained until 50 days after harvest. Decomposition rates of the biodegradable mulching films at 112 days after soybean transplanting ranged from 9.8 to 26.7% in 2022 and 13 to 36% in 2023. Although soil pH and EC varied based on the year and timing of measurements, there was no significant difference between areas that used biodegradable mulching films and PE films. Soil organic matter, nitrate and exchangeable cation contents such as Ca, Mg, and K were not significantly different in areas that used both PE films and biodegradable films. However, significantly higher levels of available phosphoric acid content were measured in areas that used biodegradable mulch films E, S, and T. Regardless of which films were used, there were no significant differences in the soil's physical properties. In 2022 and 2023, there was no difference between areas that used biodegradable mulch films and PE films. However, soil temperature in mulched areas was 2℃ higher and soil moisture was 5-15% higher than in non-mulched areas. Barley growth was not affected by being planted in soil that had been used for soybean cultivation with biodegradable films. Therefore, the biodegradable mulch films used in this study can be used without negatively affecting the growth, yield, and soil environment of soybeans.

• The Mathematical Education
• /
• v.63 no.2
• /
• pp.123-138
• /
• 2024

This study described the utilization of ChatGPT in teaching and students' learning processes for the course "Introductory Mathematics for Artificial Intelligence (Math4AI)" at 'S' University. We developed a customized ChatGPT and presented a learning model in which students supplement their knowledge of the topic at hand by utilizing this model. More specifically, first, students learn the concepts and questions of the course textbook by themselves. Then, for any question they are unsure of, students may submit any questions (keywords or open problem numbers from the textbook) to our own ChatGPT at https://math4ai.solgitmath.com/ to get help. Notably, we optimized ChatGPT and minimized inaccurate information by fully utilizing various types of data related to the subject, such as textbooks, labs, discussion records, and codes at http://matrix.skku.ac.kr/Math4AI-ChatGPT/. In this model, when students have questions while studying the textbook by themselves, they can ask mathematical concepts, keywords, theorems, examples, and problems in natural language through the ChatGPT interface. Our customized ChatGPT then provides the relevant terms, concepts, and sample answers based on previous students' discussions and/or samples of Python or R code that have been used in the discussion. Furthermore, by providing students with real-time, optimized advice based on their level, we can provide personalized education not only for the Math4AI course, but also for any other courses in college math education. The present study, which incorporates our ChatGPT model into the teaching and learning process in the course, shows promising applicability of AI technology to other college math courses (for instance, calculus, linear algebra, discrete mathematics, engineering mathematics, and basic statistics) and in K-12 math education as well as the Lifespan Learning and Continuing Education.

• Annals of Hepato-Biliary-Pancreatic Surgery
• /
• v.28 no.1
• /
• pp.70-79
• /
• 2024

Backgrounds/Aims: After pancreatoduodenectomy (PD), an early oral diet is recommended; however, the postoperative nutritional management of PD patients is known to be highly variable, with some centers still routinely providing parenteral nutrition (PN). Some patients who receive PN experience clinically significant complications, underscoring its judicious use. Using a large cohort, this study aimed to determine the proportion of PD patients who received postoperative nutritional support (NS), describe the nature of this support, and investigate whether receiving PN correlated with adverse perioperative outcomes. Methods: Data were extracted from the Recurrence After Whipple's study, a retrospective multicenter study of PD outcomes. Results: In total, 1,323 patients (89%) had data on their postoperative NS status available. Of these, 45% received postoperative NS, which was "enteral only," "parenteral only," and "enteral and parenteral" in 44%, 35%, and 21% of cases, respectively. Body mass index < 18.5 kg/m² (p = 0.03), absence of preoperative biliary stenting (p = 0.009), and serum albumin < 36 g/L (p = 0.009) all correlated with receiving postoperative NS. Among those who did not develop a serious postoperative complication, i.e., those who had a relatively uneventful recovery, 20% received PN. Conclusions: A considerable number of patients who had an uneventful recovery received PN. PN is not without risk, and should be reserved for those who are unable to take an oral diet. PD patients should undergo pre- and postoperative assessment by nutrition professionals to ensure they are managed appropriately, and to optimize perioperative outcomes.

• Food Science and Preservation
• /
• v.31 no.3
• /
• pp.423-432
• /
• 2024

In this study, a vacuum tumbler with 4 impellers (DVT) was designed and applied for thawing frozen pork (vacuum -60 kPa, jacket 35℃, 1 rpm). Quality characteristics of the thawed pork were compared with those of industrially thawed meat by natural air at room temperature (NAT) and imported vacuum tumbler (IVT). The thawing time for frozen pork (303.36 kg) using DVT (165 min) was much shorter than that of NAT (4,200 min). DVT-thawed pork had lower drip loss (0.85%) than NAT (2.08%). DVT-thawed pork showed a pH of 5.92, a total bacterial count of 1.96±0.02 log CFU/g and no coliforms. Deteriorations in fat (TBARS 0.31±0.01 MDA mg/kg) and protein (VBN 5.67±1.98 mg%) in DVT-thawed pork were significantly lower than those of NAT (p<0.05). DVT-thawed pork had a high water-holding capacity (WHC, 97.5%). The hardness (34.59±0.46 N) and chewiness (188.21±0.17) of cooked DVT-thawed pork were about 5-6 times lower than those of NTA. Microstructure (SEM) showed myofibrillar damage in NAT-thawed pork, whereas dense myofibrillar structure was observed in DVT-thawed pork. DVT was better or similar to IVT in all evaluation parameters. The designed DVT is expected to be used as an efficient thawing method in terms of processing time and yield and to produce thawed meat with high WHC, soft texture, and low spoilage by minimizing tissue damage.

• Journal of Bio-Environment Control
• /
• v.32 no.4
• /
• pp.434-441
• /
• 2023

Determining the size or area of a plant's leaves is an important factor in predicting plant growth and improving the productivity of indoor farms. In this study, we developed a convolutional neural network (CNN)-based model to accurately predict the length and width of lettuce leaves using photographs of the leaves. A callback function was applied to overcome data limitations and overfitting problems, and K-fold cross-validation was used to improve the generalization ability of the model. In addition, ImageDataGenerator function was used to increase the diversity of training data through data augmentation. To compare model performance, we evaluated pre-trained models such as VGG16, Resnet152, and NASNetMobile. As a result, NASNetMobile showed the highest performance, especially in width prediction, with an R_squared value of 0.9436, and RMSE of 0.5659. In length prediction, the R_squared value was 0.9537, and RMSE of 0.8713. The optimized model adopted the NASNetMobile architecture, the RMSprop optimization tool, the MSE loss functions, and the ELU activation functions. The training time of the model averaged 73 minutes per Epoch, and it took the model an average of 0.29 seconds to process a single lettuce leaf photo. In this study, we developed a CNN-based model to predict the leaf length and leaf width of plants in indoor farms, which is expected to enable rapid and accurate assessment of plant growth status by simply taking images. It is also expected to contribute to increasing the productivity and resource efficiency of farms by taking appropriate agricultural measures such as adjusting nutrient solution in real time.

• Journal of Bio-Environment Control
• /
• v.33 no.1
• /
• pp.71-78
• /
• 2024

Among the various environmental conditions necessary for growing crops, light is closely related to the anthesis. This study aimed to determine the optimal photoperiod affecting floral differentiation in an edible flower, marigold, to efficiently cultivate the crops in a closed-type plant factory. The experiment was conducted with photoperiods of 4, 8, 12, and 16 hours. French marigold (Tagetes patula L.) 'Durango Red' seeds were sown in polyurethane sponges, and the photoperiod treatments were applied immediately. The extent of floral differentiation was examined at 2-3 day intervals, defined as the visible appearance of flower buds at least 2 mm in size. The growth parameters such as shoot fresh weight and dry weight, height, and leaf area were measured. The optimal photoperiod was determined based on the days when the floral differentiation had occurred in 50% of the total plants. In the 4-hour treatment, proper growth and flower buds did not appear. From the 8-hour treatment, the plant grew normally, and floral differentiation occurred, however, the 8-hour treatment showed the slowest floral differentiation compared to the 12 hours treatments or more. The 12- and 16-hour treatments didn't show significant differences in floral differentiation. While the 16-hour treatment exhibited the highest results in all growth parameters, it was not significantly different from the 12-hour treatment except for shoot dry weight and leaf area. According to the results, 8 hours of photoperiod induced floral differentiation. However, more time was required for flower bud formation, and plant growth was significantly lower compared to photoperiods of 12 hours or more. Considering the energy consumption and its growth, the optimal photoperiod for marigold was 12 hours.

• Journal of Food Hygiene and Safety
• /
• v.39 no.4
• /
• pp.335-342
• /
• 2024

This study aimed to examine the delivery conditions and microbial contamination of fresh-cut and ready-to-eat foods purchased from online markets between February and November 2023. Upon arrival, the average surface temperature of the products was 11.3℃. In the fresh-cut foods, the average number of total aerobic bacteria and coliforms was 4.5 log colony-forming units (CFU)/g and 1.2 log CFU/g, respectively, whereas in the ready-to-eat foods, these values were 10.6 log CFU/g and 1.2 log CFU/g, respectively. Pathogens, such as Staphylococcus aureus, Salmonella spp., Clostridium perfringens, Listeria monocytogenes, and pathogenic Escherichia coli were absent from all samples. Bacillus cereus was found in 2.7% of the fresh-cut foods and 0.9% of the ready-to-eat foods, with contamination levels averaging 0.05 log CFU/g and 0.01 log CFU/g, respectively. In the four samples in which B. cereus was detected, genetic testing of the six toxin genes produced by B. cereus revealed the presence of at least one enterotoxin gene, excluding the emetic toxin. L. monocytogenes was absent from ready-to-eat foods but was detected in 0.9% of fresh-cut foods. Analysis of the isolated L. monocytogenes confirmed the presence of six pathogenicity-related genes, including iap, indicating the potential risk of foodborne diseases.