• International journal of advanced smart convergence
• /
• v.12 no.1
• /
• pp.18-30
• /
• 2023

We compared empirically the forecast accuracies of the LSTM model, and the ARIMA model. ARIMA model used auto.arima function. Data used in the model is 100 days. We compared with the forecast results for 50 days. We collected the stock closing prices of the top 4 companies by market capitalization in Korea such as "Samsung Electronics", and "LG Energy", "SK Hynix", "Samsung Bio". The collection period is from June 17, 2022, to January 20, 2023. The paired t-test is used to compare the accuracy of forecasts by the two methods because conditions are same. The null hypothesis that the accuracy of the two methods for the four stock closing prices were the same were rejected at the significance level of 5%. Graphs and boxplots confirmed the results of the hypothesis tests. The accuracies of ARIMA are higher than those of LSTM for four cases. For closing stock price of Samsung Electronics, the mean difference of error between ARIMA and LSTM is -370.11, which is 0.618% of the average of the closing stock price. For closing stock price of LG Energy, the mean difference is -4143.298 which is 0.809% of the average of the closing stock price. For closing stock price of SK Hynix, the mean difference is -830.7269 which is 1.00% of the average of the closing stock price. For closing stock price of Samsung Bio, the mean difference is -4143.298 which is 0.809% of the average of the closing stock price. The auto.arima function was used to find the ARIMA model, but other methods are worth considering in future studies. And more efforts are needed to find parameters that provide an optimal model in LSTM.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.1
• /
• pp.1-11
• /
• 2023

Deep-learning-based image segmentation is one of the most widely employed lane detection approaches, and it requires a post-process for extracting the key points on the lanes. A general approach for key-point extraction is using a fixed threshold defined by a user. However, finding the best threshold is a manual process requiring much effort, and the best one can differ depending on the target data set (or an image). We propose a novel key-point extraction algorithm that automatically adapts to the target image without any manual threshold setting. In our adaptive key-point extraction algorithm, we propose a line-level normalization method to distinguish the lane region from the background clearly. Then, we extract a representative key point for each lane at a line (row of an image) using a kernel density estimation. To check the benefits of our approach, we applied our method to two lane-detection data sets, including TuSimple and CULane. As a result, our method achieved up to 1.80%p and 17.27% better results than using a fixed threshold in the perspectives of accuracy and distance error between the ground truth key-point and the predicted point.

• Korean Chemical Engineering Research
• /
• v.61 no.2
• /
• pp.226-233
• /
• 2023

To create an environmentally sustainable fuel with a low sulfur concentration, requires alternative sulfur removal methods. During the course of this study, a high surface gamma alumina-supported ZnO nanocatalyst with a ZnO/-Al₂O₃ ratio of 12% was developed and tested for its ability to improve the activity of the oxidative desulfurization (ODS) process for the desulfurization of kerosene fuel. Scanning electron microscopy (SEM) and Brunauer-Emmett-Teller (BET) were used to characterize the produced nanocatalyst. In a digital batch baffled reactor (20~80 min), the effectiveness of the synthesized nanocatalyst was tested at different initial concentrations of dibenzothiophene (DBT) of 300~600 ppm, oxidation temperatures (25~70 ℃), and oxidation periods (0.5, 1, and 2 hours). The baffles included in the digital baffled batch reactor resist the swirling of the reaction mixture, thus facilitating mixing. The ODS procedure yielded the maximum DBT conversion (95.5%) at 70 ℃ with an 80-minute reaction time and an initial DBT level of 600 ppm. The most precise values of kinetic variables were subsequently determined using a mathematical modelling procedure for the ODS procedure. The average absolute error of the simulation findings was less than 5%, demonstrating a good degree of agreement with the experimental results acquired from all runs. The optimization of the operating conditions revealed that 99.1% of the DBT can be removed in 140 minutes.

• Journal of dental hygiene science
• /
• v.22 no.1
• /
• pp.37-43
• /
• 2022

Background: Among the health problems in old age, oral health is closely related to nutrition intake and digestion, so although it is an important factor in the well-being of the elderly along with general health, studies examining the relationship between oral health-related factors and grip strength of the elderly are insufficient. Therefore, this study intends to examine the relationship between oral health-related factors and grip strength, which are closely related to the general health of the elderly. Methods: This study used data from the 7th period of Korea National Health and Nutrition Survey (2016~2018) approved by the Research Ethics Review Committee of the Korea Centers for Disease Control and Prevention. Complex sample frequency analysis and descriptive statistics were performed, and general linear model analysis was performed to confirm the relationship between demographic characteristics, oral health -related factors and grip strength. The statistical analysis was performed using IBM SPSS Statistics for Windows, Version 23.0, and the significance test was based on type I error level of 0.05. Results: Grip strength was higher in the case of no discomfort than in the case of discomfort in relation to mastication discomfort and grip strength (B=0.927, p<0.001). In addition, the grip strength was decreased by 1.348 times when not using dental floss (p<0.001) and when not using mouth wash was 1.480 times (p<0.001). Conclusion: In this study, in the relationship between oral health-related factors and grip strength, grip strength was found to be high in the absence of mastication discomfort. and in the case of using dental floss and mouthwash the elderly showed high grip strength. Therefore, it is suggested to present a lifestyle to improve hand function and grip strength in the elderly and develop a program to increase grip strength and provide them at the same time during oral health education.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.161-161
• /
• 2021

본 연구에서는 금강 상류의 용담댐 유역(930.0 km²)을 대상으로 Sentinel-1 SAR(Synthetic Aperture Radar) 및 Sentinel-2 MultiSpectral Instrument(MSI) 위성영상을 활용한 토양수분 산출연구를 수행하였다. 연구에 사용된 자료는 10 m 해상도의 Sentinel-1 IW(Interferometric Wide swath) mode GRD(Ground Range Detected) product의 VV(Vertical transmit-Vertical receive) 및 VH(Vertical transmit-Horizontal receive) 편파자료와 Sentinel-2 Level-2A Bottom of Atmosphere(BOA) reflectance 자료를 2019년에 대해 각 6일 및 5일 간격으로 구축하였다. 위성영상의 Image processing은 SNAP(SentiNel Application Platform)을 활용하여 Sentinel-1 영상의 편파 별(VV, VH) 후방산란계수와 Sentinel-2의 적색(Band-4) 및 근적외(Band-8) 영상을 생성하였다. 토양수분 산출 모형은 다중선형회귀모형(Multiple Linear Regression Model)을 활용하였으며, 각 지점에 해당하는 토양 속성별로 모형을 생성하였다. 모형의 입력자료는 Sentinel-1 위성의 편파별 후방산란계수, Sentinel-1 위성에서 산출된 식생지수 RVI(Radar Vegetation Index)와 Sentinel-2 위성에서 산출된 NDVI(Normalized Difference Vegetation Index)를 활용하여 식생의 영향을 반영하고자 하였다. 모의 된 토양수분을 검증하기 위해 6개 지점의 TDR(Time Domain Reflectometry) 기반 실측 토양수분 자료를 수집하고, 상관계수(Correlation Coefficient, R), 평균제곱근오차(Root Mean Square Error, RMSE) 및 IOA(Index of Agreement)를 활용하여 전체 기간 및 계절별로 나누어 검증할 예정이다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.335-337
• /
• 2021

Flood damage is occurring all over the world, and the number of people living in flood-prone areas reached 86 million, a 25% increase compared to 2000. These floods cause enormous damage to life and property, and it is essential to decide on an appropriate evacuation in order to reduce the damage. Evacuation in anticipation of a flood also incurs a lot of cost, and if an evacuation is not performed due to an error in the flood prediction, a greater cost is incurred. Therefore, in this paper, we propose a flood risk determination model using the CNN model to enable evacuation at an appropriate time by using the time series data of precipitation and water level. Through this, it is thought that it can be utilized as an initial study to determine the time of flood evacuation to prevent unnecessary evacuation and to ensure that evacuation can be carried out at an appropriate time.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.21 no.1
• /
• pp.273-286
• /
• 2022

An HD map is essential in the automated driving of level 4 and above to generate the vehicle's global path since it contains road information and each road's lane information. Therefore, all the road elements in the HD map must be correctly defined to construct the correct road network necessary to generate the global path. But unfortunately, it is not difficult to find various errors even in the most recent HD maps. Hence, a preprocessing algorithm has been developed to detect and correct errors in the HD map. This error detection and correction result in constructing the correct road network for use in global path planning. Furthermore, the algorithm was tested on real roads' HD maps, demonstrating its validity.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.449-449
• /
• 2022

제주도는 용수의 대부분을 지하수에 의존하므로 지하수위의 예측 및 관리는 매우 중요한 사항이다. 제주도의 지층은 화산활동에 의한 현무암이 겹겹이 쌓여있는 형태를 나타내며 육지의 지층구조와 매우 다른 복잡한 형태를 나타낸다. 이에 따라 제주도 지하수위의 예측은 매우 난해하며, 최근에는 딥러닝 인공지능 모델을 활용하여 지하수위를 예측하는 연구사례가 증가하고 있다. 기존의 연구들은 인공지능 모델들이 지하수위를 적절히 예측한다고 보고하고 있으나 예측의 적절성에 대한 판단기준을 제시하지 못하였으므로 이에 대한 명확한 제시가 필요하다. 본 연구의 목표는 인공지능을 활용한 지하수위 예측오차가 허용 가능한지 판단할 수 있는 기준을 제시함에 있다. 이를 위해 전 세계의 과거 20년 동안 관련 연구결과들을 수집 및 분석하였으며, 분석 결과 인공지능 모델의 지하수위 예측오차는 지하수위 변동성이 큰 지역일수록 증가하는 것을 확인하였다. 이것은 지하수위의 변동형태가 크고 복잡할수록 인공지능 모델의 지하수위 예측성능은 낮아진다는 것을 의미한다. 이 관계를 명확하게 나타내기 위해 지하수위 최대변동폭과 평균제곱근오차 및 최대오차와의 관계를 선형회귀식으로 도출하여 허용가능한 예측오차 기준을 제시하였다. 그리고 기존 연구들에서 제시한 Nash-Sutcliffe 효율성지수와 결정계수를 분석하여 선형회귀식에 의한 기준을 보완할 수 있는 추가적인 기준을 제시하였다. 본 연구에서 제시한 인공지능 모델에 의한 지하수위 예측결과의 적절성 판단기준은 향후 지속적으로 증가하는 인공지능 예측연구에 유용하게 사용될 수 있다.

• Smart Media Journal
• /
• v.11 no.7
• /
• pp.52-60
• /
• 2022

Due to global warming and changes in the natural environment, flood damage to agricultural land due to wind and flood damage continues. Although disaster prevention projects have been continuously carried out since the founding of the country, progress has been insufficient compared to the sustained period, and huge costs are still being consumed. Therefore, it is necessary to use predictive simulation for pre-emptive response to inundation of farmland. In this paper, a case of immersion analysis simulation using a GIS(Geospatial Information System) based SWMM model was introduced, and the validity was confirmed through the error rate between our simulation result and the results of other models in the US and Korea. In addition, in the direction of using the simulation for agricultural land inundation, we presented various utilization methods to supplement the current agricultural land inundation-based information policy, such as the creation of flood traces. If simulation results from more regions are accumulated in the form of the flood analysis maps in the future, it is expected that they will be able to be utilized in various applications for preemptive response to and prevention of water disasters at the national level.

• International Area Studies Review
• /
• v.15 no.1
• /
• pp.107-133
• /
• 2011

This paper aims to analyze the causal relationships between agricultural exports and economic growth in the U.S. economy by 50 states. Using the annual data from 1973 to 2007, the theoretical methodologies based on the export-led growth (ELG) model under the static model, the impulse response function (IRF) and forecast error variation decomposition (FEVD) under the vector autoregressive (VAR) model, and the Granger causality test. The results show the causal relationship between agricultural exports and economic growth at the states' level. Especially, the ELG hypothesis is strongly supported in the case of 16 states (HI, ID, KS, MD, MI, MN, NJ, NC, ND, OK, OR, RI, SD, TX, WA, and WI) and is also weakly supported in the case of 31 states. Therefore, the agricultural exports are important factor of developing in the U.S. economy, and furthermore some states (located in coastal area and breadbasket) indicate the strong evidence for agricultural exports-led growth.