• Journal of Biosystems Engineering
• /
• v.8 no.1
• /
• pp.61-69
• /
• 1983

A system dynamic model was developed to predict food grain production under the dynamic consideration of the production circumstance and inputs such as farm population, investment on agriculture, arable land, extensive technology and weather. By using the model, the variation of the food grain production from 1978 to 2008 was examined. The results of the model output says it is desirable that the persistent and long-term program should be studied to get necessary food grain production under the variational inputs and circumstances.

• Spatial Information Research
• /
• v.15 no.3
• /
• pp.267-278
• /
• 2007

Internationally many models are developed and applied to predict the impact of the climate change, as occurring a lot of symptoms by climate change. Also, in Korea, according to increasing the application of the climate effect model in many research fields, it is required to study the method for preparing climate data and the characteristics of the climate. In this study IDSW (Inverse Distance Squared Weighting), one of the spatial statistic methods, is applied to interpolate. This method estimates a point of interest by assigning more weight to closer points, which are limited to be select by 3 in 100 km radius. As a result, annual average temperature and precipitation had increased by $0.4^{\circ}C$ and 412 mm during 1977 to 2006. They are also predicted to increase by $3.96^{\circ}C$, 319 mm in the 2100 compared to 2007. High variability of temperature and precipitation for last 30 years shows in some part of the Gangwon-do and in the southern part of Korea. Besides in the study of the variable trend, the variability of temperature and precipitation is inclined to increase in Gangwon-do and southern east part, respectively. However, during 2071 to 2100 variability of temperature is predicted to be high in midwest of Korea and variability of precipitation in the east. In the trend of variability, variability of temperature is apt to increase into west south, and variability of precipitation increase in midwest and a part of south.

• Journal of the Korean Society of Radiology
• /
• v.5 no.1
• /
• pp.19-25
• /
• 2011

Photoplethysmogram(PPG) is the method to obtain the biomedical signal using the linear relationships between the blood volume for changing the cardiac contraction and relaxation and the amount of light for absorbing the hemoglobin in the blood. In this paper, we proposed the analyzed results which show the heart rate variability and the distribution of heart rate for before and after using PPG. Moreover, this paper designed and implemented the system based on personal computer to predict cardiovascular disease in advance using the analyzed results for the autonomic balance from taking the spectral analysis of heart rate and the state of the blood vessel for analyzing APG(acceleration plethysmogram).

• Journal of Internet Computing and Services
• /
• v.23 no.3
• /
• pp.125-135
• /
• 2022

This study aims to explore variables using machine learning and provide analysis techniques suitable for predicting pharmacy sales whether government statistical indicators built to create an industrial ecosystem based on data, network, and artificial intelligence affect pharmacy sales. Therefore, this study explored predictive variables and performance through machine learning techniques such as Random Forest, XGBoost, LightGBM, and CatBoost using analysis data from January 2016 to December 2021 for 28 government statistical indicators and pharmacies in the retail sector. As a result of the analysis, economic sentiment index, economic accompanying index circulation change, and consumer sentiment index, which are economic indicators, were found to be important variables affecting pharmacy sales. As a result of examining the indicators MAE, MSE, and RMSE for regression performance, random forests showed the best performance than XGBoost, LightGBM, and CatBoost. Therefore, this study presented variables and optimal machine learning techniques that affect pharmacy sales based on machine learning results, and proposed several implications and follow-up studies.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.392-392
• /
• 2022

세계적 규모의 팬데믹 감염병의 출현은 전 세계적으로 경제적, 문화적, 사회적 파급효과가 매우 강력하며 전 인류를 위협하고 있다. 최근에 발병한 중증급성 호흡기질환 코로나바이러스 2(Severe Acute Respiratory Syndrome Coronavirus 2, SARS-CoV-2)는 2019년 12월 중국 우한에서 첫 보고 되었고 2022년 현재까지 종식되지 않고 있으며 바이러스의 전파력과 치명률이 높고 무증상 감염상태일 때에도 전염이 가능하여 현재 역학조사의 사후적 대응에 대한 한계가 있어 선제적 대응을 위한 수단이 필수 불가결해지고 있는 실정이다. 하수기반역학(Waste Based Epidemiology, WBE)이란 하수처리장으로 유입되기 전의 하수를 분석하여 하수 집수구역 내 도시민의 생활상을 예측하는 것으로 하수로 배출된 감염자의 분비물 및 배설물 속 바이러스를 하수관로에서 신속하게 검출함으로써 특정지역의 감염성 질환 전파 정도와 유행하는 타입(변이)등을 분석하고 기존 역학조사의 문제점을 극복할 수 있으며 선제적인 대응이 가능하다. 현재 COVID-19의 대유행과 관련하여 WBE를 기반으로 한 다양한 연구가 진행되고 있으며 실제 환자의 발생과 상관관계가 있음이 확인되고 있고 백신 접종과 새롭게 발생한 변이바이러스의 관계 속에서 발생하는 변수를 고려한 모델이 없다는 점을 들어 새로운 감염병 확산 예측 모델에 대한 필요성 또한 커지고 있다. 본 연구에서는 병원에서부터 하수처리장까지의 하수관거와 하수처리장에서의 SARS-CoV-2 검출농도 및 거동을 파악하는 것을 목적으로 하고 있으며 COVID-19의 감염규모 확산에 관한 방법론에서 수학적모델 (Euler Method, RK4 Method, Gillespie Algorithm)과 딥러닝 기반의 Nowcasting model과 Fore casting model을 살펴보고자 한다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.1
• /
• pp.51-59
• /
• 2021

The deep learning model can produce false prediction results due to inputs that deviate from training data through variation, which leads to fatal accidents in areas such as autonomous driving and security. To ensure reliability of the model, the model's coping ability for exceptional situations should be verified through various mutations. However, previous studies were carried out on limited scope of models and used several mutation types without separating them. Based on the CIFAR10 data set, widely used dataset for deep learning verification, this study carries out reliability verification for total of six models including various commercialized models and their additional versions. To this end, six types of input mutation algorithms that may occur in real life are applied individually with their various parameters to the dataset to compare the accuracy of the models for each of them to rigorously identify vulnerabilities of the models associated with a particular mutation type.

• KDI Journal of Economic Policy
• /
• v.22 no.1_2
• /
• pp.193-283
• /
• 2000

외환위기 이후 재정적자가 급격히 확대되면서 재정적자에 대한 일반인들의 관심이 높아지고 있다. 그러나 재정적자가 거시경제에 구체적으로 어떤 영향을 미치는가에 대한 실증분석은 많지 않은 편이다. 본고는 재정적자가 민간저축률과 물가상승률에 미치는 영향을 살펴보고 있다. 본 논문의 결과를 요약하면 다음과 같다. 첫째, 저축률과 재정적자 사이에는 리카도 동등가설이 예측하는 것과 같은 관계가 표면적으로 발견된다. 즉, 재정지출이 변하지 않을 때 재정적자의 증가는 민간저축률을 증가시켜 국민경제 전체의 저축률은 크게 변하지 않는다. 둘째, 재정수지가 변하지 않더라도 재정지출의 증가는 민간저축을 감소시킨다. 그리고 재정수지가 변하든 변하지 않든 정부소비나 이전지출의 증가는 국민저축률을 감소시킨다. 셋째, 재정적자는 물가에 별 영향을 주지 않는다. 이 가운데 첫째와 셋째의 결과는 별로 새삼스러운 것이 되지 못한다. 그러나 둘째의 결과는 지금까지 논의되지 않았던 사실을 알려주고 있다. 특히 1980년대 말 이후 GDP 대비 재정규모가 추세적으로 증가하고 있으며, 최근의 외환위기 이후에는 금융구조조정 지원 등에 따라 재정규모가 급격히 증가하고 있고, 장기적으로는 국민연금급여 등 사회보장지출의 증가가 예상됨을 고려할 때, 재정규모 증가를 억제하는 일에 보다 적극적인 노력을 기울일 필요가 있음을 알게 된다. 한편 본고에서의 한국은행의 준(準)재정활동을 고려하지 않았으나, 이를 고려할 때에도 재정수지가 물가상승률에 별다른 영향을 미치지 않는지에 대한 추가적 연구가 필요하다고 판단된다.

• Journal of Korean Society of Transportation
• /
• v.27 no.6
• /
• pp.139-146
• /
• 2009

While air pollutants emission caused by the traffic is one of the major sources, few researches have done. This study investigated the extent to which traffic and road related characteristics such as traffic volumes, speeds and road weather data including wind speed, temperature and humidity, as well as the road geometry affect the air pollutant emission. We collected the real time air pollutant emission data from Seoul automatic stations and real time traffic volume counts as well as the road geometry. The regression air pollutant emission models were estimated. The results show followings. First, the more traffic volume increase, the more pollutant emission increase. The more vehicle speed increase, the more measurement quantity of pollutant decrease. Secondly, as the wind speed, temperature, and humidity increase, the amount of air pollutant is likely to decrease. Thirdly, the figure of intersections affects air pollutant emission. To verify the estimated models, we compared the estimates of the air pollutant emission with the real emission data. The result show the estimated results of Chunggae 4 station has the most reliable data compared with the others. This study is differentiated in the way the model used the real time air pollutant emission data and real time traffic data as well as the road geometry to explain the effects of the traffic and road characteristics on air quality.

• Information Systems Review
• /
• v.23 no.1
• /
• pp.23-43
• /
• 2021

Recently, many studies have been conducted to increase the accuracy of stock price prediction by analyzing candlestick charts using artificial intelligence techniques. However, these studies failed to consider the time-series characteristics of candlestick charts and to take into account the emotional state of market participants in data learning for stock price prediction. In order to overcome these limitations, this study produced input data by combining volatility index and candlestick charts to consider the emotional state of market participants, and used the data as input for a new method proposed on the basis of combining variantion autoencoder (VAE) and attention mechanisms for considering the time-series characteristics of candlestick chart. Fifty firms were randomly selected from the S&P 500 index and their stock prices were predicted to evaluate the performance of the method compared with existing ones such as convolutional neural network (CNN) or long-short term memory (LSTM). The results indicated the method proposed in this study showed superior performance compared to the existing ones. This study implied that the accuracy of stock price prediction could be improved by considering the emotional state of market participants and the time-series characteristics of the candlestick chart.

• Journal of the Korean Geographical Society
• /
• v.50 no.3
• /
• pp.277-287
• /
• 2015

This paper addresses limitations of land-change modeling application in the context of REDD (Reducing Emissions from Deforestation and forest Degradation). REDD is an international conservation policy that aims to protect forests via carbon credit generation and trading. In REDD, carbon credits are generated only if there is measurable quantied carbon sequestration activities that are additional to business-as-usual (BAU). A "reference level" is defined as simulated baseline carbon emissions for the future under a BAU scenario, and predictive land-change modeling plays an important role in constructing reference levels. It is tested in this research how predictive accuracies of two land-change models, namely Geographic Emission Benchmark (GEB) and GEOMOD, vary with respect to different spatial scales: Xishuangbanna prefecture and Yunnan province. The accuracies are measured by Figure of Merit. In this Chinese case study, it turns out that GEB's better performance is mainly due to quantity (e.g., how many hectares of forest will be converted to agricultural land?) rather than spatial allocation (e.g., where will the conversion happen?). As both quantity and allocation are crucial in REDD reference level setting it appears to be fundamental to systematically analyze accuracies of quantity and allocation independently in pursuit of accurate reference levels.