• 한국측량학회지
• /
• 제34권4호
• /
• pp.383-390
• /
• 2016

Remote sensing data has been widely used in the estimation of crop yields by employing statistical methods such as regression model. Machine learning, which is an efficient empirical method for classification and prediction, is another approach to crop yield estimation. This paper described the corn yield estimation in Iowa State using four machine learning approaches such as SVM (Support Vector Machine), RF (Random Forest), ERT (Extremely Randomized Trees) and DL (Deep Learning). Also, comparisons of the validation statistics among them were presented. To examine the seasonal sensitivities of the corn yields, three period groups were set up: (1) MJJAS (May to September), (2) JA (July and August) and (3) OC (optimal combination of month). In overall, the DL method showed the highest accuracies in terms of the correlation coefficient for the three period groups. The accuracies were relatively favorable in the OC group, which indicates the optimal combination of month can be significant in statistical modeling of crop yields. The differences between our predictions and USDA (United States Department of Agriculture) statistics were about 6-8 %, which shows the machine learning approaches can be a viable option for crop yield modeling. In particular, the DL showed more stable results by overcoming the overfitting problem of generic machine learning methods.

• 한국농공학회지
• /
• 제40권1호
• /
• pp.68-77
• /
• 1998

The purpose of this paper is to establish a method estimating the daily urban water demand using statistical analysis that is used for developing the efficient management and operation of the water supply facilities, and accurary of the model is verified by error rate and F-value. The data used in this study were the daily urban water use, the weather conditions such as temperature, precipitation, relative humidity, etc, and the day of The week. The case study was taken placed for the city of Namwon in Korea. The raw data used in this study were rearranged either by month or by season for analysis purpose, and the statistical analysis was applied to the data to obtain the regression model As a result of this study, the linear regression model was developed to estimate the daily urban water use with weather condition. The regression constant and coefficients of the model were determined for each month of a year. The accuracy of the model was within 3% of average error and within 11% of maximum error. The resulting model was found to he useful to the practical operation and management of the water supply facilities.

• 한국연초학회지
• /
• 제11권1호
• /
• pp.27-39
• /
• 1989

This study was conducted to determine the time and methods of predicting tobacco yield. by analysis of climatic factors in the period of tobacco season during 8 years from 1979 to 1986 at the Daegu district, south eastern part of Korean peninsular. The results obtained are summarised as follows: 1. Climatic factors of each month which have influence on tobacco yield were the amount of rainfall in May and sunshine hours in July. Among climatic factors at tobacco growth stages, the precipitation yield. But these meteorological factors had different effect on variety. 2. Between tobacco yields and climatic factors by even values of each month, tobacco yield was estimated by equations, flue cured tobacco :Y=190.6-5.230X1+ 0.474$\times$2 + 0.142X3(Xl : Minimum temperature of April, X2: Precipitation during May, X3:Sunshine duration on July), air cured tobacco : Y= 195.3-0.447Xl + 0.363$\times$2 + 0.l12$\times$3(Xl :Maximum temperature of May, X2:Precipitation during May. X3: Sunshine duration on July). While between tobacco yield and climatic factors at different growth stage, predicting equation of yield could be derived, flue cured tobacco : Y=205.8+0.510Xl +0.289$\times$2 + 0.305$\times$3 (Xl :Average temperature during the early growth stage, X2 :Precipitation during the early and maximum growth stage, X3 : Sunshine hours during the leaf and tips maturing stage), air cured tobacco Y=194.T-0.498Xl 10.615$\times$2+0.121$\times$3(Xl ;Maximum temperature during the transplanting time, X2 : Precipitation during the maximum growth stage, X3 : Sunshine hours during the leaf and tips maturing stage).

• 대한의용생체공학회:의공학회지
• /
• 제27권6호
• /
• pp.370-375
• /
• 2006

The purpose of this work is to predict the systolic blood pressure (BP) during exercise from pulse transit time (PTT) for warning of possible danger. PTT was calculated as the time between R-peak of ECG and the peak of differential photoplethysmograph (PPG). For the PTT-BP model, we used regress equations from previous studies and 3 kinds of new models combining linear and nonlinear regress equation. The model parameters were estimated with the data measured under low to middle intensity exercise, and then was tested with the data measured under high intensity exercise. Predicted BP values after high intensity exercise were compared with those measured by cuff-type sphygmomanometer. The results showed that the error between measured and predicted values were acceptable for the monitoring BP. We tested PTT-BP models 1 month after the identification without further calibration. Models could predict the BP and the errors between measured and predicted BP were about 5mmHg. The suggested system is expected to be helpful in recognizing any danger during exercise.

• 한국보건간호학회지
• /
• 제16권1호
• /
• pp.201-212
• /
• 2002

This study was designed to identify the risk factors of unplanned readmission in a university hospital. The six-month discharge information from January to June, 2000 in a tertiary university hospital was used as a source of data through the medical record and hospital information system. To increase the effect of comparison. the data were collected by sampling 192 couples (384 patients) of unplanned readmission group through the matching by its disease groups, sex, and age. The accuracy of prediction for unplanned readmission was analyzed by constructing the predicted model of unplanned readmission through the logistic regression. The study results are as follows. The conditional logistic regression analysis was performed with nine variables at the significance level 0.05 through univariate analysis including residence, days after discharge, initial admission route, previous admission, transfer to special care unite, hospital stay days, medical care expenses, special cares, and laboratory and imaging services. As a result, the closer the patients live in Seoul and Gyeong-in area (Odds ratio=2.529, p=0.003), the shorter the days after discharge was (Odds ratio=0.600, p=0.000), and the more frequent admission rate was (Odds ratio=2.317, p=0.004), the more unplanned readmission was resulted. Also, the accuracy of prediction for data classification of this regression model showed $70.3\%$(032+83/306).

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2015년도 학술발표회
• /
• pp.224-224
• /
• 2015

The East Asia monsoon is one of the most complex atmospheric phenomena caused by Land-Sea thermal contrast. It plays essential role in fulfilling the water needs of the region but also poses stern consequences in terms of flooding and droughts. This study analyzed the influence of Global Sea Surface Temperature Anomalies (SSTA) on occurrence of droughts in the East Asia monsoon region ($20^{\circ}N-50^{\circ}N$, $103^{\circ}E-149^{\circ}E$). Standardized Precipitation Index (SPI) was employed to characterize the droughts over the region using 30-year (1978-2007) gridded rainfall dataset at $0.5^{\circ}$ grid resolution. Due to high variability in intensity and spatial extent of monsoon rainfall the East Asia monsoon region was divided into the homogeneous rainfall zones using cluster analysis method. Seven zones were delineated that showed unique rainfall regimes over the region. The influence of SSTA was assessed by using lagged-correlation between global gridded SSTA ($0.2^{\circ}$ grid resolution) and SPI of each zone. Sea regions with potential influence on droughts in different zones were identified based on significant positive and negative correlation between SSTA and SPI with a lag period of 3-month. The results showed that SSTA have the potential to be used as predictor variables for prediction of droughts with a reasonable lead time. The findings of this study will assist to improve the drought prediction over the region.

• 한국농공학회논문집
• /
• 제49권4호
• /
• pp.3-12
• /
• 2007

The purpose of this paper is to analyze the relationship between meteorological factors and agricultural reservoir storage, and predict the reservoir storage by multiple regression equation selected by high correlated meteorological factors. Two agricultural reservoirs (Geumgwang and Gosam) located in the upsteam of Gongdo water level gauging station of Anseong-cheon watershed were selected. Monthly reservoir storage data and meteorological data in Suwon weather station of 21 years (1985-2005) were collected. Three cases of correlation (case 1: yearly mean, case 2: seasonal mean dividing a year into 3 periods, and case 3: lagging the reservoir storage from 1 month to 3 months under the condition of case 2) were examined using 8 meteorological factors (precipitation, mean/maximum/minimum temperature, relative humidity, sunshine hour, wind velocity and evaporation). From the correlation analysis, 4 high correlated meteorological factors were selected, and multiple regression was executed for each case. The determination coefficient ($R^{2}$) of predicted reservoir storage for case 1 showed 0.45 and 0.49 for Geumgwang and Gosam reservoir respectively. The predicted reservoir storage for case 2 showed the highest $R^{2}$ of 0.46 and 0.56 respectively in the period of April to June. The predicted reservoir storage for 1 month lag of case 3 showed the $R^{2}$ of 0.68 and 0.85 respectively for the period of April to June. The results showed that the status of agricultural reservoir storage could be expressed with couple of meteorological factors. The prediction enhanced when the storage data are divided into periods rather than yearly mean and especially from the beginning time of paddy irrigation (April) to high decrease of reservoir storage (June) before Jangma.

• 한국수자원학회논문집
• /
• 제38권6호
• /
• pp.471-483
• /
• 2005

일단위 강우-유출모형인 SSARR모형을 이용하여 한강, 낙동강, 섬진강유역에 월 앙상블 유량예측 시스템을 구축하였다. 우선 SSARR모형의 월 평균 유출량에 대한 모의정확성을 평가한 결과 한강과 낙동강유역에서는 과소추정하는 경향이 뚜렷하였으며, 섬진강유역에서는 모의오차의 분산이 커 정확성 개선이 필요하였다. 최적선형 보정기법을 적용하여 SSARR모형의 모의유량을 보정한 결과, 섬진강을 제외한 한강과 낙동강유역의 검증지점에서는 모의 정확성이 크게 개선되었다. 또한 1998년부터 2003년까지 월 앙상블 유량예측을 실시하여 예측 정확성을 평가하였다. 한강과 낙동강유역에서 최적선형 보정기법을 이용할 경우 앙상블 유량예측 정확성이 크게 개선되었으나, 섬진강유역은 개선효과가 미비하였다.

• 대한한의학회지
• /
• 제41권2호
• /
• pp.58-79
• /
• 2020

Objectives: The purpose of this study is to predict the weight loss by applying machine learning using real-world clinical data from overweight and obese adults on weight loss program in 4 Korean Medicine obesity clinics. Methods: From January, 2017 to May, 2019, we collected data from overweight and obese adults (BMI≥23 kg/m2) who registered for a 3-month Gamitaeeumjowi-tang prescription program. Predictive analysis was conducted at the time of three prescriptions, and the expected reduced rate and reduced weight at the next order of prescription were predicted as binary classification (classification benchmark: highest quartile, median, lowest quartile). For the median, further analysis was conducted after using the variable selection method. The data set for each analysis was 25,988 in the first, 6,304 in the second, and 833 in the third. 5-fold cross validation was used to prevent overfitting. Results: Prediction accuracy was increased from 1^st to 2^nd and 3^rd analysis. After selecting the variables based on the median, artificial neural network showed the highest accuracy in 1^st (54.69%), 2^nd (73.52%), and 3^rd (81.88%) prediction analysis based on reduced rate. The prediction performance was additionally confirmed through AUC, Random Forest showed the highest in 1^st (0.640), 2^nd (0.816), and 3^rd (0.939) prediction analysis based on reduced weight. Conclusions: The prediction of weight loss by applying machine learning showed that the accuracy was improved by using the initial weight loss information. There is a possibility that it can be used to screen patients who need intensive intervention when expected weight loss is low.

• Journal of Chest Surgery
• /
• 제55권2호
• /
• pp.143-150
• /
• 2022

Background: The effectiveness of extracorporeal membrane oxygenation (ECMO) for patients with refractory cardiogenic shock or cardiac arrest is being established, and serum lactate is well known as a biomarker of end-organ perfusion. We evaluated the efficacy of pre-ECMO lactate for predicting 6-month survival in patients with acute coronary syndrome (ACS) undergoing ECMO. Methods: We reviewed the medical records of 148 patients who underwent veno-arterial (VA) ECMO for ACS between January 2015 and June 2020. These patients were divided into survivors and non-survivors based on 6-month survival. All clinical data before and during ECMO were compared between the 2 groups. Results: Patients' mean age was 66.0±10.5 years, and 116 (78.4%) were men. The total survival rate was 45.9% (n=68). Cox regression analysis showed that the pre-ECMO lactate level was an independent predictor of 6-month mortality (hazard ratio, 1.210; 95% confidence interval [CI], 1.064-1.376; p=0.004). The area under the receiver operating characteristic curve of pre-ECMO lactate was 0.64 (95% CI, 0.56-0.72; p=0.002; cut-off value=9.8 mmol/L). Kaplan-Meier survival analysis showed that the cumulative survival rate at 6 months was significantly higher among patients with a pre-ECMO lactate level of 9.8 mmol/L or less than among those with a level exceeding 9.8 mmol/L (57.3% vs. 31.8%, p=0.0008). Conclusion: A pre-ECMO lactate of 9.8 mmol/L or less may predict a favorable outcome at 6 months in ACS patients undergoing VA-ECMO. Further research aiming to improve the accuracy of predictions of reversibility in patients with high pre-ECMO lactate levels is essential.