• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.234-234
• /
• 2023

유해 남조 대발생(Harmful Algal blooms, HABs)이 담수호에 발생하면 마이크로시스틴과 같은 독성물질과 맛·냄새 물질을 생성하여 상수원이용과 친수활동을 방해한다. 그래서 유해 남조 대발생 전 유해남조 세포수를 예측하여 선제적 대응하는 것은 중요하다. 따라서 본 연구는 머신러닝기반 Random Forest(RF)를 활용하여 팔당댐 앞의 유해남조 세포수를 예측하는 모델을 개발하고 성능을 평가하고자 한다. 모델 구축을 위해 2012년 4월부터 2021년 12월까지의 팔당호(삼봉리, 경안천) 및 남북한강(의암댐~이포보)권역의 조류, 수질, 수리/수문, 기상 자료를 수집하여 입력 및 출력 자료로 이용하였다. 수집된 데이터에는 다양한 입력변수들이 있어 남조 세포수 예측 성능 비교를 위한 전체 26개 변수 적용과 통계학적으로 상관관계가 높은 12개 변수 적용을 통해 모델을 구축하였다. 입력, 출력 자료로 이용한 유해남조 세포수는 로그변환된 값으로 사용하였으며 일반적인 조류 시료 채취기간이 7일이므로 7일 후를 예측하기 위한 모델을 구축하였다. 구축한 모델의 성능은 실측데이터와 예측데이터의 R²로 산출하여 평가하였다. 전체 26개 입력변수로 모델 구축 후 학습 및 검증 수행 결과 R²의 학습 0.803, 검증 0.729로 나타났고, 유해남조 세포수와 유의미한 상관관계를 보이는 12개 입력변수로 모델 구축 후 학습 및 검증 수행 R²은 학습 0.784, 검증 0.731로 나타났다. 두 모델의 성능을 살펴본 결과 입력변수 개수의 변화에 따른 성능차이는 크지 않은 것으로 나타났으며, 남조세포수 예측을 위한 모델로서 활용가능함을 알 수 있었다. 향후 연구에서는 Random Forest 외 다른 기계학습 모델들과 딥러닝 모델을 통해 남조세포수 예측 성능이 높은 모델을 구축해볼 필요성이 있다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.7
• /
• pp.956-962
• /
• 2020

Recently, as the size of data used in an embedded system increases, the need for an ECC decoding operation to robustly receive a massive data is emphasized. In this paper, we propose a method to accelerate the execution of computations that derive syndrome vectors when ECC decoding is performed using Hamming code in an embedded system with a built-in GPU. The proposed acceleration method uses the matrix-vector multiplication of the decoding operation using the CSR format, one of the data structures representing sparse matrix, and is performed in parallel in the CUDA kernel of the GPU. We evaluated the proposed method using a target embedded board with a GPU, and the result shows that the execution time is reduced when ECC decoding operation accelerated based on the GPU than used only CPU.

• Journal of Korea Entertainment Industry Association
• /
• v.14 no.8
• /
• pp.427-436
• /
• 2020

This study aims to understand human rights sensibility in university varsity teams and to compare and analyze differences in human rights sensibility by variables related to university varsity teams. In order to achieve the purpose of this study, 188 student-athletes from 10 universities were selected. The collected data were analyzed in descriptive analysis, reliability analysis, one-way ANOVA, Scheffe using IBM SPSS 24.0. First, based on the episodes, the episodes with the highest human rights sensitivity are the right to labor of migrant workers and happiness rights, and the episodes with the lowest human rights sensitivity are the right to freedom of detention and privacy rights. In addition, among the sub-factors of human rights sensitivity, perception of responsibility and perception of behavior are higher than perception of outcome. Second, there are differences in the human rights sensitivity of the university varsity team depending on the size and the level of performance of the university varsity team. Third, there are differences in the human rights sensibility of the university varsity team depending on the educational characteristics (volunteer activity and human rights education) of the university varsity team.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.32 no.6
• /
• pp.587-601
• /
• 2020

The ocean circulation was simulated in the East Sea and Ulleungdo-Dokdo region using ROMS (Regional Ocean Modeling System) model. By adopting the East Sea 3 km model and the HYCOM 9 km data, Ulleungdo 1 km model and Ulleungdo-Dokdo 300 m model were constructed with one-way grid nesting method. During the model development, a correction method was proposed for the distortion of the open boundary data which may be caused by the bathymetry data difference between the mother and child models and the interpolation/extrapolation method. Using this model, a super-high resolution ocean circulation with a horizontal resolution of 300 m near the Ulleungdo and Dokdo region was simulated for year 2018. In spite of applying the same conditions except for the initial and boundary data, the numerical models result indicated significantly different characteristics in the study area. Therefore, these results were compared and verified by using the surface current data estimated by satellites altimeter data and temperature data from NIFS (National Institute of Fisheries Science). They suggest that in general, the improvement of the one-way grid nesting with the HYCOM data on RMSE, Mean Bias, Pattern correlation and Vector correlation is greater in 300 m model than in the 1 km model. However, the nesting results of using East Sea 3 km model showed that simulations of the 1 km model were better than 300 m model. The models better resolved distinct ridge/trough structures of isotherms in the vertical sections of water temperature when using the higher horizontal resolution. Furthermore, Karman vortex street was simulated in Ulleungdo-Dokdo 300 m model due to the terrain effect of th islands that was not shown in the Ulleungdo 1 km model.

• Korean Journal of Remote Sensing
• /
• v.30 no.6
• /
• pp.755-767
• /
• 2014

In this study, detailed air flow characteristics in an urban areas were analyzed using GIS data and a Computational Fluid Dynamics (CFD) model. For this, a building construction algorithm optimized for Geographic Information System (GIS) data with a vector format (Los Angeles region imagery acquisition consortium 2 geographic information system, LARIAC2 GIS) was used. In the LARIAC2 GIS data, building vertices were expressed as latitude and longitude. Using the model buildings constructed by the algorithm as the surface boundary data in the CFD model, we performed numerical simulations for two building-congested areas in Los Angeles using inflow information provided by California Air Resources Board. Comparing with the inflow, there was a marked difference in wind speed and direction within the target areas, which was mainly caused by the secondarily induced local circulations such as street-canyon vortices, horse-shoe vortices, and recirculation zones. In street canyons parallel to the inflow direction, wind speed increased due to a channeling effect and, in street canyons perpendicular to the inflow direction, vertically well developed vortices were induced. In front of a building, a horse-shoe vortex was developed near the surface and, behind a building, a recirculation zone was developed. Near the surface in the areas where the secondarily induced local circulations, wind speed remarkably increased. Overall, wind direction little (largely) changed at the areas where wind speed largely increased (decreased).

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.27 no.2
• /
• pp.225-234
• /
• 2009

Real-time Aerial Monitoring System (RAMS) is to perform the rapid mapping in an emergency situation so that the geoinformation such as orthophoto and/or Digital Elevation Model is constructed in near real time. In this system, the GPS/INS plays an very important role in providing the position as well as the attitude information. Therefore, in this study, the performance of an IMU sensor which is supposed to be installed on board the RAMS is evaluated. And the integration algorithm of GPS/INS are tested with simulated dataset to find out which is more appropriate in real time mapping. According to the static and kinematic results, the sensor shows the position error of 3$\sim$4m and 2$\sim$3m, respectively. Also, it was verified that the sensor performs better on the attitude when the magnetic field sensor are used in the Aerospace mode. In the comparison of EKF and UKF, the overall performances shows not much differences in straight as well as in curved trajectory. However, the calculation time in EKF was appeared about 25 times faster than that of UKF, thus EKF seems to be the better selection in RAMS.

• Journal of Korean Society for Geospatial Information Science
• /
• v.24 no.4
• /
• pp.83-88
• /
• 2016

The use of consistent coastlines is an important element for the systematic management of maritime boundaries and the interests of local governments. The Hydrographic and Oceanographic Agency conducted a preliminary survey for consistent coastline production, since 2001. As a result, the length of coastline was different by year. Because of the lack of systematic management, the use of incorrect data, etc. We also changed the coastline on the sea chart to show on a digital map for realization of terrain expression method. However, there was a variation in shoreline length due to various surveying techniques and shoreline extraction methods. In this paper, the characteristics of Jeju-do coastline were analysed by using a modified divider method of fractal dimension. The accuracy of the vectorization was determined by converting the actual distance in the Public Survey Amendment for proper divider use. With 1:5,000 and 1:25,000 digital maps of Jeju-si and Seogwipo-si each fractal dimensions were calculated. Jeju-si=1.14 and Seogwipo-si=1.12 in 1: 5,000. Jeju-si=1.13 and Seogwipo-si=1.10 in 1: 25,000. Calculated fractal dimension were correlated to data from digital maps. It was considered that complexity and scale of coastlines affected. In the future coastline length statistics and minimum ratio of calculated coastline length to original length need to be determined for consistency of coastline length statistics.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.25 no.1
• /
• pp.48-59
• /
• 2023

Ground-level ozone affects human health and plant growth. Ozone is produced by chemical reactions between oxides of nitrogen (NOx) and volatile organic compounds (VOCs) from anthropogenic and biogenic sources. In this study, two different land cover and emission factor datasets were input to the MEGAN v2.1 emission model to examine how these parameters contribute to the biogenic emissions and ozone production. Four input sensitivity scenarios (A, B, C and D) were generated from land cover and vegetation emission factors combination. The effects of BVOCs emissions by scenario were also investigated. From air quality modeling result using CAMx, maximum 1 hour ozone concentrations were estimated 62 ppb, 60 ppb, 68 ppb, 65 ppb, 55 ppb for scenarios A, B, C, D and E, respectively. For maximum 8 hour ozone concentration, 57 ppb, 56 ppb, 63 ppb, 60 ppb, and 53 ppb were estimated by scenario. The minimum difference by land cover was up to 25 ppb and by emission factor that was up to 35 ppb. From the modeling performance evaluation using ground ozone measurement over the six regions (East Seoul, West Seoul, Incheon, Namyangju, Wonju, and Daegu), the model performed well in terms of the correlation coefficient (0.6 to 0.82). For the 4 urban regions (East Seoul, West Seoul, Incheon, and Namyangju), ozone simulations were not quite sensitive to the change of BVOC emissions. For rural regions (Wonju and Daegu) , however, BVOC emission affected ozone concentration much more than previously mentioned regions, especially in case of scenario C. This implies the importance of biogenic emissions on ozone production over the sub-urban to rural regions.

• Archives of design research
• /
• v.18 no.2 s.60
• /
• pp.219-232
• /
• 2005

Moderns are living within flood of web contents, animation, reflex data etc. as well as sight, product, environment design. There fore, modern consumer has much options. Designer must provide various result for consumer for this reason. And must invent new sensitivity and propose to consumer. As purpose of this MCC sensitivity palette research takes advantage of the most sensitive color, do. Because applying correct sensitivity more than when design with matter already settled, rid private prejudice, and is thing to convey design intention exactly to user. Excellent culture contents must be able to equip international color design sensitivity. MCC sensitivity palette research studies and carries on the head emotion and sensitivity language that is nationality first, and collect End arranged sensitivity adjective through data analysis and picture data analysis that is the next time research leader Munheonjeok. And distributed collected adjective equally, and arrange distributed adjective by field of each sensitivity and collect system. Do 3 colors, 4 colors color scheme in selected sensitivity adjective and completed Simheom version. Result of beta version research to color specialist and designer last digital palette through question and inquiry compose. Through this process, completed more real and correct digital color sensitivity palette. Completed color scheme is operated in www.mcdri.net on web, and also programs to windows base and developed to software. MCC color scheme palette that research result is made includes sensitivity data database. This database can use directly in industry and continuous development is available. Software can search color scheme in language and idea development through classification search that use 3 attributes of color is available there is cough data of each output device different color error.

• The Korean Journal of Applied Statistics
• /
• v.31 no.3
• /
• pp.367-382
• /
• 2018

Prediction models for a corporate bond rating in existing studies have been developed using various models such as linear regression, ordered logit, and random forest. Financial characteristics help build prediction models that are expected to be contained in the assigning model of the bond rating agencies. However, the ranges of bond ratings in existing studies vary from 5 to 20 and the prediction models were developed with samples in which the target companies and the observation periods are different. Thus, a simple comparison of the prediction accuracies in each study cannot determine the best prediction model. In order to conduct a fair comparison, this study has collected corporate bond ratings and financial characteristics from 2013 to 2017 and applied prediction models to them. In addition, we applied the elastic-net penalty for the linear regression, the ordered logit, and the ordered probit. Our comparison shows that data-driven variable selection using the elastic-net improves prediction accuracy in each corresponding model, and that the random forest is the most appropriate model in terms of prediction accuracy, which obtains 69.6% accuracy of the exact rating prediction on average from the 5-fold cross validation.