• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.28 no.4
• /
• pp.133-142
• /
• 2023

In this paper, we built a data-driven artificial intelligence model using RNN-LSTM (Recurrent Neural Networks-Long Short-Term Memory) to predict the Lorenz system, and examined the possibility of whether this model can replace chaotic dynamic models. We confirmed that the data-driven model reflects the chaotic nature of the Lorenz system, where a small error in the initial conditions produces fundamentally different results, and the system moves around two stable poles, repeating the transition process, the characteristic of "deterministic non-periodic flow", and simulates the bifurcation phenomenon. We also demonstrated the advantage of adjusting integration time intervals to reduce computational resources in data-driven models. Thus, we anticipate expanding the applicability of data-driven artificial intelligence models through future research on refining data-driven models and data assimilation techniques for data-driven models.

• Korean Journal of Culture and Social Issue
• /
• v.16 no.2
• /
• pp.103-124
• /
• 2010

The purpose of the present study was to examine the relationships among private tutoring hours, academic motivation, use of learning strategies, and academic achievement test scores using structural equation modeling. The sample consisted of 3,607 7th graders from Korean middle schools who were included in the Korean Education Longitudinal Study. The results suggest that there was no evidence that the private tutoring hours predicted students' motivation and learning strategy use. It was found that the private tutoring hours predicted achievements in English and Math, but it was negligible in magnitude. As for achievement test scores, academic motivation and the use of learning strategies played more critical role rather than the private tutoring hours.

• Information Systems Review
• /
• v.25 no.3
• /
• pp.99-120
• /
• 2023

Courier services users' experience of violating privacy affects psychology and behavior of protecting personal privacy. Depending on what privacy infringement experience (PIE) of courier services users, learning about perceived privacy infringement incidents is made, recognition is formed, affection is formed, and behavior is appeared. This paradigm of changing in privacy psychologies of courier services users has an important impact on predicting responses of privacy protective action (PPA). In this study, a theoretical research framework are developed to explain the privacy protective action (PPA) of courier services users by applying attitude theory. Based on this framework, the relationships among past privacy infringement experience (PIE), perceived privacy risk (PPR), privacy concerns (i.e., concerns in unlicensed secondary use (CIUSU), concerns in information error (CIE), concerns in improper access (CIA), and concern in information collection (CIC), and privacy protective action (PPA) are analyzed. In this study, the proposed research model was surveyed by people with experience in using courier services and was analyzed for finding relationships among research variables using structured an equation modeling software, SMART-PLS. The empirical results show the causal relationships among PIE, PPR, privacy concerns (CIUSU, CIE, CIA, and CIC), and PPA. The results of this study provide useful theoretical implications for privacy management research in courier services, and practical implications for the development of courier services business model.

• KOREAN JOURNAL OF PACKAGING SCIENCE & TECHNOLOGY
• /
• v.29 no.3
• /
• pp.145-151
• /
• 2023

In order to extend shelf life for rockfish fillets by modified atmosphere packaging (MAP), different package atmospheres were compared in the product quality preservation. Firstly, CO₂ solubility was measured at 0, 5, 10, and 15℃ to be incorporated into the mathematical model to predict the volume and CO₂ concentration of the package at expected storage temperature. The CO₂ solubility given in Henry's constant decreased with temperature to be fitted with a linear equation. Then air packaging as control and four MAP conditions of 100 g fillets were prepared and stored for duration of 5 days at 10℃ to compare them in the quality preservation effect. Four MAP conditions employed were CO₂(60):O₂(30):N₂(10), CO₂(60):O₂(0):N₂(40), CO₂(30):O₂(30):N₂(40) andCO₂(30):O₂(0):N₂(70). MAP conditions with high CO₂ concentration inhibited total aerobic bacteria, and the conditions containing O₂ led to low TVB-N. MAP of CO₂(60):O₂(30):N₂(10) was found to be the best condition for rockfish fillet preservation considering total aerobic bacteria growth and nitrogenous volatiles production.

• Journal of Oral Medicine and Pain
• /
• v.30 no.4
• /
• pp.433-446
• /
• 2005

The purpose of this study was to investigate the sociopsychological factors which might influence the course of the temporomandibular disorders and to develop models for prediction of treatment outcome related to pain, dysfunction and sound. For this study, 268 patients with temporomandibular disorders were selected from the patients presented to department of Oral Medicine, Wonkwang university dental hospital. Chief complaints of these subjects were largely grouped into three categories such as pain, dysfunction and sound, and 10 cm visual analogue scale(VAS) was used to record the state of the three complaints every visit and Treatment Index(VAS TI) was calculated from VAS. All the forty-two items obtained from clinical examination or questionnaire were statistically processed with $SPSS^{(R)}$ windows. The results of this study were as follows: 1. The items showed a difference between male and female subjects were maximum mouth opening, palpation score, jaw jerk during mandibular movement, and sleep disturbance. Among questionnaires such as SRRS, HAD scale and PSQI, the HAD scales showed the most highly significant correlation with the each scale item of the SCL-90R. 2. Among the groups classified by VAS TI, the group with the lowest VAS TI showed the highest VAS score in the start of treatment but showed the lowest score in the end of treatment, without respect to the type of chief complaint. From these results, it is assumed that the active treatment duration for the subjects with lower VAS score in the first visit would be longer than the subjects with higher score with the somewhat poor treatment outcome. 3. With regard to all the three complaints, the items showing significant effect in the model for prediction of treatment outcome were from questionnaire, except one item, maximum mouth opening, which suggest that the sociopsychological factors would be strongly related to development and progress of the symptoms.

• Korean Journal of Food Science and Technology
• /
• v.30 no.2
• /
• pp.341-347
• /
• 1998

To develop simple and accurate analytical method for freezing time prediction of beef and tylose under various freezing conditions, freezing time (Y) was regressed against the reciprocal $(X_3)$ of difference of initial freezing point and freezing medium temperature, reciprocal $(X_4)$ of surface heat transfer coefficient, the initial temperature $(X_1)$ and thickness $(X_2)$ of samples which should cover most situations arising in frozen food industry. As results of the multiple regression analysis, equations were obtained as follows. $Y_{tylose}=3.45X_1+7642.84X_2+4642.67X_3+2946.89X_4-431.33\;(R^2=0.9568)$ and $Y_{beef}=0.68X_1+7568.98X_2+2430.78X_3+3293.26X_4-299.00\;(R^2=0.9897)$. These equations offered better results than Plank, Nagaoka and Pham's models, shown in satisfactory agreement with models of Cleland & Earle and Hung & Thompson when were compared to previous models, and the accuracy of its was very high as average absolute difference of about 10% in the difference between the fitted and experimental results. Also, thermal diffusivities of beef and tylose were measured as $4.43{\times}10^{-4}m^2/hr$ and $4.39{\times}10^{-4}m^2/hr$ at $6{\sim}7^{\circ}C$, $2.42{\times}10^{-3}m^2/hr$ and $3.32{\times}10^{-3}m^2/hr$ at $-10{\sim}-12^{\circ}C$. Initial freezing points of beef and tylose were $-1.2^{\circ}C\;and\;-0.6^{\circ}C$, respectively. Surface heat transfer coefficients were estimated $20.57\;W/m^2^{\circ}C$ with no-packing, $16.11\;W/m^2^{\circ}C$ with wrap packing and $13.07\;W/m^2^{\circ}C$ with Al-foil packing, and the cooling rate of immersion freezing method was about 10 times faster than that of air blast freezing method.

• Korean Journal of Poultry Science
• /
• v.44 no.4
• /
• pp.259-265
• /
• 2017

This study was to investigate the effect of chick weight in the early growth period on market-weight of Korean native chickens (KNCs). We measured the body weights of 1,087 chickens (male 479, female 608) of 13 KNC strains at 1-84 days of age at two week intervals. The growth performance of the 13 KNC strains was investigated. Correlation coefficients among the weights of chickens in each growth period and regression of market-weight on early chick weights were analyzed. The results showed that the average body weight of 70-day-old KNCs was 1,962g: 2,154g and 1,819.7g for males and females, respectively. The equation for regression of body weight on age was estimated as $\hat{Y}=0.1347X^2+18.738X-40.134$ ($R^2=0.9418$). Using this regression equation, the duration required to attain a KNC market-weight of 2 kg was estimated as 71.8 days. All the correlation coefficients between early chick weight and market-weight were significantly positive. Although the correlation coefficients among the chick weights in each growth period decreased with increase in age interval, early growth weight had a significant effect on late growth performance. The correlation coefficient between market-weight at 70-days and chick weight at 1-day was estimated to be a low as 0.10-0.13. In the estimations of market-weight correlation coefficients, correlation coefficient and coefficient of determination were high in the female and male chicks after 28-days and 42-days of age, respectively. The results of the analysis of correlation and regression between early chick weight and market-weight of KNCs showed that market-weight could be predicted based on the weights of 28-day-old females and 42-day-old males.

• Journal of the Korean earth science society
• /
• v.36 no.1
• /
• pp.90-108
• /
• 2015

The WRF-Chem model was applied to simulate the Asian dust event affecting the Korean Peninsula from 11 to 13 November 2010. GOCART dust emission schemes, RADM2 chemical mechanism, and MADE/SORGAM aerosol scheme were adopted within the WRF-Chem model to predict dust aerosol concentrations. The results in the model simulations were identified by comparing with the weather maps, satellite images, monitoring data of $PM_{10}$ concentration, and LIDAR images. The model results showed a good agreement with the long-range transport from the dust source area such as Northeastern China and Mongolia to the Korean Peninsula. Comparison of the time series of $PM_{10}$ concentration measured at Backnungdo showed that the correlation coefficient was 0.736, and the root mean square error was $192.73{\mu}g/m^3$. The spatial distribution of $PM_{10}$ concentration using the WRF-Chem model was similar to that of the $PM_{2.5}$ which were about a half of $PM_{10}$. Also, they were much alike in those of the UM-ADAM model simulated by the Korean Meteorological Administration. Meanwhile, the spatial distributions of $PM_{10}$ concentrations during the Asian dust events had relevance to those of both the wind speed of u component ($ms^{-1}$) and the PBL height (m). We performed a regressive analysis between $PM_{10}$ concentrations and two meteorological variables (u component and PBL) in the strong dust event in autumn (CASE 1, on 11 to 23 March 2010) and the weak dust event in spring (CASE 2, on 19 to 20 March 2011), respectively.

• Journal of Bio-Environment Control
• /
• v.32 no.4
• /
• pp.384-395
• /
• 2023

Since transpiration plays a key role in optimal irrigation management, knowledge of the irrigation demand of crops like tomatoes, which are highly susceptible to water stress, is necessary. One way to determine irrigation demand is to measure transpiration, which is affected by environmental factor or growth stage. This study aimed to estimate the transpiration amount of tomatoes and find a suitable model using mathematical and deep learning models using minute-by-minute data. Pearson correlation revealed that observed environmental variables significantly correlate with crop transpiration. Inside air temperature and outside radiation positively correlated with transpiration, while humidity showed a negative correlation. Multiple Linear Regression (MLR), Polynomial Regression model, Artificial Neural Network (ANN), Long short-term Memory (LSTM), and Gated Recurrent Unit (GRU) models were built and compared their accuracies. All models showed potential in estimating transpiration with R² values ranging from 0.770 to 0.948 and RMSE of 0.495 mm/min to 1.038 mm/min in the test dataset. Deep learning models outperformed the mathematical models; the GRU demonstrated the best performance in the test data with 0.948 R² and 0.495 mm/min RMSE. The LSTM and ANN closely followed with R² values of 0.946 and 0.944, respectively, and RMSE of 0.504 m/min and 0.511, respectively. The GRU model exhibited superior performance in short-term forecasts while LSTM for long-term but requires verification using a large dataset. Compared to the FAO56 Penman-Monteith (PM) equation, PM has a lower RMSE of 0.598 mm/min than MLR and Polynomial models degrees 2 and 3 but performed least among all models in capturing variability in transpiration. Therefore, this study recommended GRU and LSTM models for short-term estimation of tomato transpiration in greenhouses.

• Korean Journal of Agricultural Science
• /
• v.16 no.1
• /
• pp.84-101
• /
• 1989

An accurate measurement of size, surface area and volume of agricultural products is essential in many engineering operations such as handling and sorting, and in heat transfer studies on heating and cooling processes. Little information is available on these properties due to their irregular shape, and moreover very little information on the rough rice, soybean, barley, and wheat has been published. Physical dimensions of grain, such as length, width, thickness, surface area, and volume vary according to the variety, environmental conditions, temperature, and moisture content. Especially, recent research has emphasized on the variation of these properties with the important factors such as moisture content. The objectives of this study were to determine physical dimensions such as length, width and thickness, surface area and volume of the rough rice, soybean, barley, and wheat as a function of moisture content, to investigate the effect of moisture content on the properties, and to develop exponential equations to predict the surface area and the volume of the grains as a function of physical dimensions. The varieties of the rough rice used in this study were Akibare, Milyang 15, Seomjin, Samkang, Chilseong, and Yongmun, as a soybean sample Jangyeobkong and Hwangkeumkong, as a barley sample Olbori and Salbori, and as a wheat sample Eunpa and Guru were selected, respectively. The physical properties of the grain samples were determined at four levels of moisture content and ten or fifteen replications were run at each moisture content level and each variety. The results of this study are summarized as follows; 1. In comparison of the surface area and the volume of the 0.0375m diameter-sphere measured in this study with the calculated values by the formula the percent error between them showed least values of 0.65% and 0.77% at the rotational degree interval of 15 degree respectively. 2. The statistical test(t-test) results of the physical properties between the types of rough rice, and between the varieties of soybean and wheat indicated that there were significant difference at the 5% level between them. 3. The physical dimensions varied linearly with the moisture content, and the ratios of length to thickness (L/T) and of width to thickness (W/T) in rough rice decreased with increase of moisture content, while increased in soybean, but uniform tendency of the ratios in barley and wheat was not shown. In all of the sample grains except Olbori, sphericity decreased with increase of moisture content. 4. Over the experimental moisture levels, the surface area and the volume were in the ranges of about $45{\sim}51{\times}10^{-6}m^2$, $25{\sim}30{\times}10^{-9}m^3$ for Japonica-type rough rice, about $42{\sim}47{\times}10^{-6}m^2$, $21{\sim}26{\times}10^{-9}m^3$ for Indica${\times}$Japonica type rough rice, about $188{\sim}200{\times}10^{-6}m^2$, $277{\sim}300{\times}10^{-9}m^3$ for Jangyeobkong, about $180{\sim}201{\times}10^{-6}m^2$, $190{\sim}253{\times}10^{-9}m^3$ for Hwangkeumkong, about $60{\sim}69{\times}10^{-6}m^2$, $36{\sim}45{\times}10^{-9}m^3$ for Covered barley, about $47{\sim}60{\times}10^{-6}m^2$, $22{\sim}28{\times}10^{-9}m^3$ for Naked barley, about $51{\sim}20{\times}10^{-6}m^2$, $23{\sim}31{\times}10^{-9}m^3$ for Eunpamill, and about $57{\sim}69{\times}10^{-6}m^2$, $27{\sim}34{\times}10^{-9}m^3$ for Gurumill, respectively. 5. The increasing rate of surface area and volume with increase of moisture content was higher in soybean than other sample grains, and that of Japonica-type was slightly higher than Indica${\times}$Japonica type in rough rice. 6. The regression equations of physical dimensions, surface area and volume were developed as a function of moisture content, the exponential equations of surface area and volume were also developed as a function of physical dimensions, and the regression equations of surface area were also developed as a function of volume in all grain samples.