• KCID journal
• /
• v.19 no.1
• /
• pp.87-96
• /
• 2012

Information of local solar radiation is essential for many field, including water resources management, crop yield estimation, crop growth model, solar energy systems and irrigation and drainage design. Unfortunately, solar radiation measurements are not easily available due to the cost and maintenance and calibration requirements of the measuring equipment and station. Therefore, it is important to elaborate methods to estimate the solar radiation based on readily available meteorological data. In this study, two empirical equations are employed to estimate daily solar radiation using Cheongju Regional Meteorological Office data. Two scenarios are considered: (a) sunshine duration data are available for a given location, or (b) only daily cloudiness index records exist. Simple linear regression with daily sunshine duration and cloudiness index as the dependent variable accounted for 91% and 80%, respectively of the variation of solar radiation(H) at 2011. Daily global solar radiation is highly correlated with sunshine duration. In order to indicate the performance of the models, the statistical test methods of the mean bias error(MBE), root mean square error(RMSE) and correlation coefficient(r) are used. Sunshine duration and cloudiness index can be easily and reliably measured and data are widely available.

• Water for future
• /
• v.29 no.2
• /
• pp.235-247
• /
• 1996

Meteorological and soil water content data measured from semiarid watersheds of Lucky Hills and Kendall during the summer rainy and winter periods were used to study the interrelationships between the controlling variables of the evapotranspiration, and to evaluate the effects of variables on daily actual evapotranspiration (ET) estimation. Simple and multiple linear regression (MLR) analyses were employed to evaluate the order of importance of the meteorological and soil water factors involved. The information gained was used for MLR model development. Theavailable energy and vapor pressure deficit were found to be the important variables to estimate actual ET (AET) for both periods and at both watersheds. Therefore, the important variables of evapotranspiration process in these semiarid watersheds appear to be simply the components of energy term in available energy and aerodynamic term in vapor pressure deficit of Penman potential evapotranspiration (PET) equation.

• The Korean Journal of Applied Statistics
• /
• v.34 no.6
• /
• pp.923-936
• /
• 2021

A large number of non-responses are occurring in the sample survey, and various methods have been developed to deal with them appropriately. In particular, the bias caused by non-ignorable non-response greatly reduces the accuracy of estimation and makes non-response processing difficult. Recently, Chung and Shin (2017, 2020) proposed an estimator that improves the accuracy of estimation using parametric super-population model and response rate model. In this study, we suggested a bias corrected non-response mean estimator using a nonparametric function generalizing the form of a parametric super-population model. We confirmed the superiority of the proposed estimator through simulation studies.

• Journal of Korean Academy of Rural Health Nursing
• /
• v.18 no.1
• /
• pp.11-18
• /
• 2023

Purpose: This study is to compare adolescents' health behaviors by city size and to propose regionally tailored health promotion. Methods: We analyzed the data from the 17th Youth Health Risk Behavior Online Survey, national widly performed in 2021. Multi-sample descriptive and linear regression analysis was performed by city size. Results: The frequency of fruit consumption in the last week was 2.88 in the rural area, which is lower than 2.98 and 3.05 in other cities (F=10.98, p<.001). The number of high-intensity physical activity days in the last week (7 days) was 2.90 days in the rural area, higher than 2.74 and 2.73 days in other cities (F=3.36, p=.038). The number of days smoking cigarettes in the last 30 days was 3.23 days in the rural area, higher than 3.08 and 3.02 days in other cities (F=3.41, p=.035). BMI was 22.01 in the rural area, which was higher than 21.57 and 21.61 in other cities (F=4.19, p=.015). Conclusion: School health offices in the rural area districts need to operate to manage lack of fruit intake, smoking, and weight management programs in association with local healthcare institutions.

• Geomechanics and Engineering
• /
• v.37 no.6
• /
• pp.539-554
• /
• 2024

This study explores development of prediction model for seismic site classification through the integration of machine learning techniques with horizontal-to-vertical spectral ratio (HVSR) methodologies. To improve model accuracy, the research employs outlier detection methods and, synthetic minority over-sampling technique (SMOTE) for data balance, and evaluates using seven machine learning models using seismic data from KiK-net. Notably, light gradient boosting method (LGBM), gradient boosting, and decision tree models exhibit improved performance when coupled with SMOTE, while Multiple linear regression (MLR) and Support vector machine (SVM) models show reduced efficacy. Outlier detection techniques significantly enhance accuracy, particularly for LGBM, gradient boosting, and voting boosting. The ensemble of LGBM with the isolation forest and SMOTE achieves the highest accuracy of 0.91, with LGBM and local outlier factor yielding the highest F1-score of 0.79. Consistently outperforming other models, LGBM proves most efficient for seismic site classification when supported by appropriate preprocessing procedures. These findings show the significance of outlier detection and data balancing for precise seismic soil classification prediction, offering insights and highlighting the potential of machine learning in optimizing site classification accuracy.

• Journal of Radiation Industry
• /
• v.17 no.4
• /
• pp.551-557
• /
• 2023

In recent years, radiation has become a socially important issue, increasing the need for accurate prediction of radiation levels. In this study, machine learning-based models such as Multiple Linear Regression (MLR), Random Forest (RF), XGBoost, and LightGBM, which predict the dose rate by time(nSv h^-1) by selecting only important variables, were used, and the correlation between temperature, humidity, cumulative precipitation, wind direction, wind speed, local air pressure, sea pressure, solar radiation, and radiation dose rate (nSv h^-1) was analyzed by collecting weather data and radiation dose rate for about 6 months in Jangseong, Jeollanam-do. As a result of the evaluation based on the RMSE (Root Mean Squared Error) and R-Squared (R-Squared coefficient of determination) scores, the RMSE of the XGBoost model was 22.92 and the R-Squared was 0.73, showing the best performance among the models used. As a result of optimizing hyperparameters of all models using the GridSearch method and comparing them by adding variables inside the measuring instrument, it was confirmed that the performance improved to 2.39 for RMSE and 0.99 for R-Squared in both XGBoost and LightGBM.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.12
• /
• pp.6097-6120
• /
• 2019

Online consumer reviews (OCRs) provide product information and recommendations especially pictures in reviews depict the true information about the product. This study investigates the influence of pictured reviews on online seller (for a particular product of a seller) rating with moderating effect of price, brand type (foreign vs local), goods type (experience vs search), and brand familiarity. Multiple robust linear regression analysis with moderation interaction and quadratic effect used to explain the relationship of the explanatory variables with the criterion variable. We collected cross-sectional data from the two most renowned Chinese online shopping platforms (B2C) of total 15,621 product links. Results show that higher number of reviews with a low ratio of picture reviews response negative effect on rating, whereas the lower number of reviews with a high ratio of picture reviews response positive effect on the rating. In overall picture in the reviews improve the online seller product rating. For the moderation effect, results show that price and brand familiarity have a positive interaction effect on the relation of pictured reviews and rating whereas experience goods have less negative effect comparing search goods. Finally, local brand has less negative interaction effect comparing foreign brand to pictured reviews and rating.

• Research in Community and Public Health Nursing
• /
• v.30 no.2
• /
• pp.206-216
• /
• 2019

Purpose: This study is to investigate the low back pain, social support, impediment in daily living activities and to identify factors affecting impediment in elderly farmer' daily living activities. Methods: The participants were 128 elderly farmers who had received nerve block. Data were collected using a structured questionnaire from February to March, 2018. They were analyzed using t-test, ANOVA, pearson's correlation coefficient, and linear multiple regression. Results: The score of low back pain was $6.27{\pm}1.69$ (10 points), that of social support $2.92{\pm}0.76$ (1~5 points), and that of impediment in activity of daily living $2.01{\pm}0.82$ (0~5 points). Factors affecting impediment in activity of daily living were found to include age (p=.017), daily hours of farm work (p<.001), fear for the nerve block (p<.001), low back pain (p<.001), and social support (p<.001); the explanatory power of these variables was 58.8%. Conclusion: This study has found the controllable factors affecting impediment in activity of daily living among the rural elderly engaging in farm work include low back pain, social support, and daily farming hours. Therefore, to reduce impediment in activity of daily living among them, it is necessary to develop nursing interventions that can improve impediment in activity of daily living through reduction of daily farming hours using local resources. It is also desirable to improve their health status by reducing low back pain, and develop and apply social supports with health education programs that fit the local resources and the needs of the rural elderly.

• Animal Bioscience
• /
• v.35 no.5
• /
• pp.778-788
• /
• 2022

Objective: The objective of the study was to understand and document socio-economic characteristics, production parameters, challenges and management practices used by Fijian households which keep indigenous chickens. Methods: A survey involving 200 households was carried out in coastal and inland communities of Fiji's wet and semi-dry ecoregions. Data on the influence of ecoregion and location of households relative to the sea on management practices, challenges and productivity of indigenous chickens were analyzed using logistic regression and general linear model of SAS software. Results: Irrespective of location relative to the sea and ecoregion, households indicated that they kept indigenous chickens for food and income generation. The Welsummer was the most (p>0.05) preferred breed. Households in the semi-dry inland communities had the largest (p<0.05) flocks compared to those in semi-dry coastal communities and the wet region. Chickens in the semi-dry region performed better (p<0.05) than those in the wet region in terms of number of clutches per year and mature live weight. Predators and feed shortages were the biggest challenges faced by households in all areas. The mongoose was ranked as the most (p>0.05) common predator followed by domestic dogs. Most households in the wet ecoregion's coastal communities housed their chickens at night, whereas communities in semi-dry ecoregion housed their chickens most of the time (p<0.05). In all regions, no households sold their chickens to commercial markets (p>0.05). Households in semi-dry ecoregion were more likely (p>0.05) to sell their chickens at the local market place. Conclusion: The productivity of local chickens in Fiji is low because of feed shortage, predators such as the mongoose and lack of market linkages.

• Journal of the Korean association of regional geographers
• /
• v.16 no.5
• /
• pp.610-622
• /
• 2010

The proportion of overweight and obese individuals in the United States has been continuously increasing up to recently. Many studies related to obesity have concentrated on jurisdictional levels of aggregation, making it very difficult to dearly illustrate at risk regions. In other words, little research has been conducted in relation to spatial patterns considering spatial dependency and heterogeneity by spatial autocorrelation models over space. In response, this research analyzes spatial patterns between overweight/obesity and risk factors, such as high blood pressure and diabetes, over space. Specifically, the Moran''s I and Geary''s C will be conducted for global and local measures. What is more, the Ordinary Least Square (OLS) linear regression and Geographically Weighted Regression methods will be applied to identify spatial dependency and spatial heterogeneity. Data provided by the Behavioral Risk Factor Surveillance System (BRFSS) have Body-Mass Index (BMI) rates, containing 4 rates of under, healthy, overweight, and obesity. In addition, high blood pressure and diabetes rates in the United States will be used as independent variables. Lastly, we are confident that this research will be beneficial for a decision maker to make a prevention plan for obesity.