• Journal of Korea Water Resources Association
• /
• v.51 no.12
• /
• pp.1207-1216
• /
• 2018

Discharge or water level predictions at tidally affected river reaches are currently still a great challenge in hydrological practices. This research aims to predict water level of the tide dominated site, Jamsu bridge in the Han River downstream. Physics-based hydrodynamic approaches are sometimes not applicable for water level prediction in such a tidal river due to uncertainty sources like rainfall forecasting data. In this study, TensorFlow deep learning framework was used to build a deep neural network based LSTM model and its applications. The LSTM model was trained based on 3 data sets having 10-min temporal resolution: Paldang dam release, Jamsu bridge water level, predicted tidal level for 6 years (2011~2016) and then predict the water level time series given the six lead times: 1, 3, 6, 9, 12, 24 hours. The optimal hyper-parameters of LSTM model were set up as follows: 6 hidden layers number, 0.01 learning rate, 3000 iterations. In addition, we changed the key parameter of LSTM model, sequence length, ranging from 1 to 6 hours to test its affect to prediction results. The LSTM model with the 1 hr sequence length led to the best performing prediction results for the all cases. In particular, it resulted in very accurate prediction: RMSE (0.065 cm) and NSE (0.99) for the 1 hr lead time prediction case. However, as the lead time became longer, the RMSE increased from 0.08 m (1 hr lead time) to 0.28 m (24 hrs lead time) and the NSE decreased from 0.99 (1 hr lead time) to 0.74 (24 hrs lead time), respectively.

• Journal of the Korea Convergence Society
• /
• v.10 no.7
• /
• pp.15-21
• /
• 2019

Research in detecting fake information gained a lot of interest after the US presidential election in 2016. Information from unknown sources are produced in the shape of news, and its rapid spread is fueled by the interest of public drawn to stimulating and interesting issues. In addition, the wide use of mass communication platforms such as social network services makes this phenomenon worse. Poynter Institute created the International Fact Checking Network (IFCN) to provide guidelines for judging the facts of skilled professionals and releasing "Code of Ethics" for fact check agencies. However, this type of approach is costly because of the large number of experts required to test authenticity of each article. Therefore, research in automated fake news detection technology that can efficiently identify it is gaining more attention. In this paper, we investigate fake news detection systems and researches that are rapidly developing, mainly thanks to recent advances in deep learning technology. In addition, we also organize shared tasks and training corpus that are released in various forms, so that researchers can easily participate in this field, which deserves a lot of research effort.

• Journal of the Korea Institute of Building Construction
• /
• v.20 no.6
• /
• pp.505-514
• /
• 2020

This research introduces the method of developing Korea BIM standard (KBIMS) architectural and structural element library and the methodology of converting KBIMS IFC property names with special characters. Diverse BIM tools are utilizing in project, however BIM library researches lack diversity on BIM tool selection. This research described the method to generate twelve categories and seven hundred and ninety-three elements library containing geometrical and numerical data in CATIA V6. KBIMS has its special property data naming systems which was the challenge inputting to ENOVIA IFC database. Three mapping methods for special naming characters had been developed and the ASCII code method was applied. In addition, the convertor prototype had been developed for searching and replacing the ASCII codes into the original KBIMS IFC property names. The methodology was verified by exporting 2,443 entities without data loss in the sample model conversion test. This research would provide a wider choice of BIM tool selection for applying KBIMS. Furthermore, the research would help on the reduction of data interoperability issues in projects. The developed library would be open to the public, however the continuous update and maintenance would be necessary.

• Journal of Biomedical Engineering Research
• /
• v.43 no.5
• /
• pp.341-352
• /
• 2022

Medical image segmentation is the most important task in radiation therapy. Especially, when segmenting medical images, the liver is one of the most difficult organs to segment because it has various shapes and is close to other organs. Therefore, automatic segmentation of the liver in computed tomography (CT) images is a difficult task. Since tumors also have low contrast in surrounding tissues, and the shape, location, size, and number of tumors vary from patient to patient, accurate tumor segmentation takes a long time. In this study, we propose a method algorithm for automatically segmenting the liver and tumor for this purpose. As an advantage of setting the boundaries of the tumor, the liver and tumor were automatically segmented from the CT image using the 2D CoordConv DeepLab V3+ model using the CoordConv layer. For tumors, only cropped liver images were used to improve accuracy. Additionally, to increase the segmentation accuracy, augmentation, preprocess, loss function, and hyperparameter were used to find optimal values. We compared the CoordConv DeepLab v3+ model using the CoordConv layer and the DeepLab V3+ model without the CoordConv layer to determine whether they affected the segmentation accuracy. The data sets used included 131 hepatic tumor segmentation (LiTS) challenge data sets (100 train sets, 16 validation sets, and 15 test sets). Additional learned data were tested using 15 clinical data from Seoul St. Mary's Hospital. The evaluation was compared with the study results learned with a two-dimensional deep learning-based model. Dice values without the CoordConv layer achieved 0.965 ± 0.01 for liver segmentation and 0.925 ± 0.04 for tumor segmentation using the LiTS data set. Results from the clinical data set achieved 0.927 ± 0.02 for liver division and 0.903 ± 0.05 for tumor division. The dice values using the CoordConv layer achieved 0.989 ± 0.02 for liver segmentation and 0.937 ± 0.07 for tumor segmentation using the LiTS data set. Results from the clinical data set achieved 0.944 ± 0.02 for liver division and 0.916 ± 0.18 for tumor division. The use of CoordConv layers improves the segmentation accuracy. The highest of the most recently published values were 0.960 and 0.749 for liver and tumor division, respectively. However, better performance was achieved with 0.989 and 0.937 results for liver and tumor, which would have been used with the algorithm proposed in this study. The algorithm proposed in this study can play a useful role in treatment planning by improving contouring accuracy and reducing time when segmentation evaluation of liver and tumor is performed. And accurate identification of liver anatomy in medical imaging applications, such as surgical planning, as well as radiotherapy, which can leverage the findings of this study, can help clinical evaluation of the risks and benefits of liver intervention.

• Journal of Life Science
• /
• v.32 no.1
• /
• pp.36-43
• /
• 2022

Mucosal immunity is a well-designed defense system that builds precise and dynamic relationships against pathogens, and the gastrointestinal tract is the most important organ with this system, acting as a guardian at the forefront of its activity. Salmonella spp. cause food poisoning, entering the body orally and mainly invading the Peyer's patches of the small intestine. Although Salmonella strains share similar mechanisms for inducing innate immunity, different serotypes may have different effects on the intestinal mucosa due to host specificities and pathogenicity. In this study, we evaluated the effects of Salmonella enteritidis infections in mouse intestine and observed significantly reduced dose-dependent survival rates in a challenge test. Flow cytometry data showed no significant differences in intestinal immune cell populations, although histology indicated increased mucin production and decreased goblet cell counts in the Salmonella-treated groups. Furthermore, Claudin expression was significantly decreased in the samples with Salmonella. To investigate the relationship between S. enteritidis infection and inflammatory response, dextran sodium sulfate (DSS) was administered after infection and the results indicate lower survival rate after DSS treatment. In conclusion, we were able to identify the optimal concentration of S. enteritidis to modulate the intestinal mucosal immunity of mice and inflammatory response.

• Journal of Conservation Science
• /
• v.36 no.6
• /
• pp.541-548
• /
• 2020

There is a growing demand for natural materials to replace adhesives based on volatile organic compounds (VOCs). However, the exclusion of VOCs from the manufacturing process leads to difficulties in manufacturing, and reduction in productivity and preservability. In this paper, we report the manufacture of natural bioadhesives using the carrageenan component of seaweed. λ-carrageenan, isolated from the extracted total carrageenan, was used to prepare a highly stable adhesive for paper. The resulting composition was 52.0 ± 1.0% λ-carrageenan, 30.5 ± 0.5% Polyvinylpyrrolidone, 1.0 ± 0.05% ethylhexylglycerin, 1.5 ± 0.05% glycerin, 13.5 ± 0.5% dextrine, and 0.6 ± 0.05% food-grade antifoam emulsion. The viscosity was found to be 1.13 ± 0.07 × 105 cP (25℃), UV degradation occurred at pH6.22, drying rate was 15min, △b^* was -10.79, and △E^* ab was 8.18. The bioadhesive showed an excellent adhesion strength of 44.63 kgf/cm². Thus this adhesive showed excellent fungal resistance and good adhesive persistence, without the presence of total volatile organic compounds (TVOC), formaldehyde (HCHO), and heavy metals.

• Journal of fish pathology
• /
• v.35 no.1
• /
• pp.27-40
• /
• 2022

Yellow head virus (YHV), Infectious hypodermal and hematopoietic necrosis (IHHNV), Taura syndrome virus (TSV), and Infectious myositis virus (IMNV) cause serious mortality to Penaeidae shrimp in the aquaculture. In this study, YHV, IHHNV, TSV, and IMNV were surveyed from imported frozen shrimps between 2019 and 2020 via molecular diagnostic assay. Among 10 shrimp groups, YHV (n=1) and IHHNV (n=4) were detected by RT-PCR and PCR, respectively. From the phylogenetic analysis based on the partial ORF 1b region of YHV, YHV was classified into YHV genotype 3 (YHV3). And IHHNVs (n=2) detected from Litopenaeus vannamei belong to infectious IHHNV type 2. Although IHHNVs (n=2) identified from Penaeus monodon showed PCR positive results (MG 831F/R primer set), the sequences of ORF 2 and 3 were not amplified, suggesting that those samples might possess type A IHHNV related sequence of P. monodon. Furthermore, in the challenge test, even though PCR-detected isolates (YHV3/type A IHHNV related sequence or infectious IHHNV type 2) were not induced mortality to L. vannamei, viral genes were amplified suggesting that the viruses in the frozen shrimp could be non-pathogenic particles which are not enough to induce mortality.

• Journal of Korea Entertainment Industry Association
• /
• v.13 no.4
• /
• pp.207-216
• /
• 2019

The purpose of this study was to analyze the relationships among soccer media involvement experience(SMIE), purchase intent(PI) and continued participation intent(CPI) in soccer clubs and provide basic data necessary for the sustainable growth of soccer clubs and related goods companies. In order to accomplish such study purposes, the study employed a survey method with a total of 327 amateur soccer players residing in G metropolitan city. The data from the survey questionnaires were validated through exploratory factor analysis and reliability test. The data were analyzed through descriptive statistics, correlation analysis and multiple regression analysis at the significance level of .05. Accordingly, following findings were derived from the current study. First, the level of interest was the highest among SMIE factors, followed by challenge and technology. Second, the level of alternative evaluation was the highest among PI factors, followed by purchase recognition, problem recognition, and information search. Third, technology factor in SMIE had a negative effect on PI. Fourth, technology factor in SMIE had a positive effect on CPI. Lastly, among PI factors, problem recognition had a negative effect and alternative evaluation had a positive effect on CPI.

• Korean Journal of Culture and Social Issue
• /
• v.17 no.2
• /
• pp.241-260
• /
• 2011

This study was to develop the fun perception scale measuring what conditions employees experience fun feeling in work and to examine the construct validity of it. For this, the pre-studies(open-ended questionnaire, in-depth interviews, literature survey, pre-survey) were conducted to develop the preliminary questions of fun scale. In main study, 250 employees(male: 125, female, 125) were responded to a questionnaire consisted of 40 questions of fun perception scale extracted by pre-studies. The results were as follow. First, through the item analysis, factor analysis and reliability analysis, 7 factors composed of 29 items were extracted: 'self-determination', 'extrinsic reward', 'goal achievement', 'pleasure in the process', 'contribution to the company', 'worthless', 'challenge'. Cronbach's alpha reliability coefficient of each factor was suitable. Secondly, EFA was conducted to test the construct validity of fun scale with AMOS 16.0. Several goodness of fit indexes were used to assess model fit: X2/df, TLI, CFI, RMSEA. The results were revealed that all the indexes were acceptable with no additional modification. Based on these findings, the theoretical and the practical implications of fun perception scale were discussed.

• Journal of the Korea Institute of Building Construction
• /
• v.24 no.2
• /
• pp.203-213
• /
• 2024

This study delves into the mechanical properties of fiber-reinforced cement composites(FRCC) concerning the dispersion method of multi-walled carbon nanotubes(MWCNTs). MWCNTs find utility in industrial applications, particularly in magnetic sensing and crack detection, owing to their diverse properties including heat resistance and chemical stability. However, current research endeavors are increasingly directed towards leveraging the electrical properties of MWCNTs for self-sensing and smart sensor development. Notably, achieving uniform dispersion of MWCNTs poses a challenge due to variations in researchers' skills and equipment, with excessive dispersion potentially leading to deterioration in mechanical performance. To address these challenges, this study employs ultrasonic dispersion for a defined duration along with PCE surfactant, known for its efficacy in dispersion. Test specimens of FRCC are prepared and subjected to strength, drawing, and direct tensile tests to evaluate their mechanical properties. Additionally, the influence of MWCNT dispersion efficiency on the enhancement of FRCC mechanical performance is scrutinized across different dispersion methods.