• Journal of Korea Water Resources Association
• /
• v.50 no.11
• /
• pp.781-793
• /
• 2017

In this study, applicabilities of aerodynamic approaches for the estimation of pan evaporation were evaluated on 56 study stations in South Korea. To accomplish this study purpose, previous researchers' evaporation estimation equations based on aerodynamic approaches were grouped into seven generalized evaporation models. Furthermore, four multiple linear regression (MLR) models were developed and tested. The independent variables of MLR models are meteorological variables such as wind speed, vapor pressure deficit, air temperature, and atmospheric pressure. These meteorological variables are required for the application of aerodynamic approaches. In order to consider the effect of autocorrelation, MLR models were developed after differencing variables. The applicability of MLR models with differenced variables was compared with that of MLR models with undifferenced variables and the comparison results showed no significant difference between the two methods. The study results have indicated that there is strong correlation between estimated pan evaporation (using aerodynamic models and MLR models) and measured pan evaporation. However, pan evaporation are overestimated during August, September, October, November, and December. Most of meteorological variables that are used for MLR models show statistical significance in the estimation of pan evaporation. Vapor pressure deficit was turned out to be the most significant meteorological variable. The second most significant variable was air temperature; wind speed was the third most significant variable, followed by atmospheric pressure.

• Korean Journal of Culture and Social Issue
• /
• v.16 no.4
• /
• pp.447-468
• /
• 2010

The purpose of this study was to attain a clearer understanding of the reciprocal effects of deviant self-concept and delinquent behaviors by applying a latent state-trait autoregressive modeling approach. Although traditional autoregressive cross-lagged (ARCL) modeling has been widely applied to test the longitudinal reciprocal relationship between the two constructs, it could produce misspecified findings if there were trait-like processes involved in this relationship. The latent state-trait autoregressive(LST-AR) modeling was applied to control trait effects of deviant self-concept and to examine the reciprocal causal relations between the two constructs. Data were taken from a sample of 3,449 eighth graders who were followed annually for 5 years from the Korea Youth Panel Study. The combining LST-AR model with ARCL model substantiated the reciprocal effects of deviant self-concept and delinquent behaviors, even after the stable trait component of deviant self-concept was taken into account. The present findings shed lights on the reciprocal effects of behaviors (i.e., delinquency) and self concepts (i.e., deviant self-concept). Not only did behaviors change corresponding self-concept, but the ways adolescents perceived themselves influenced their behaviors.

• Tunnel and Underground Space
• /
• v.33 no.6
• /
• pp.594-609
• /
• 2023

TBM (Tunnel Boring Machine) method is gaining popularity in urban and underwater tunneling projects due to its ability to ensure excavation face stability and minimize environmental impact. Among the prominent models for predicting disc cutter life, the NTNU model uses the Cutter Life Index(CLI) as a key parameter, but the complexity of testing procedures and rarity of equipment make measurement challenging. In this study, CLI was predicted using multiple linear regression analysis and tree-based machine learning techniques, utilizing rock properties. Through literature review, a database including rock uniaxial compressive strength, Brazilian tensile strength, equivalent quartz content, and Cerchar abrasivity index was built, and derived variables were added. The multiple linear regression analysis selected input variables based on statistical significance and multicollinearity, while the machine learning prediction model chose variables based on their importance. Dividing the data into 80% for training and 20% for testing, a comparative analysis of the predictive performance was conducted, and XGBoost was identified as the optimal model. The validity of the multiple linear regression and XGBoost models derived in this study was confirmed by comparing their predictive performance with prior research.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.2
• /
• pp.299-304
• /
• 2023

Among artificial intelligence techniques, deep learning is a model that has been used in many places and has proven its effectiveness. However, deep learning models are not used effectively in everywhere. In this paper, we will show the limitations of deep learning models through comparison of regression analysis and deep learning models, and present a guide for effective use of deep learning models. In addition, among various techniques used for optimization of deep learning models, data normalization and data shuffling techniques, which are widely used, are compared and evaluated based on actual data to provide guidelines for increasing the accuracy and value of deep learning models.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.41 no.2
• /
• pp.123-131
• /
• 2021

The buffer is a key component of an engineered barrier system that safeguards the disposal of high-level radioactive waste. Buffers are located between disposal canisters and host rock, and they can restrain the release of radionuclides and protect canisters from the inflow of ground water. Since considerable heat is released from a disposal canister to the surrounding buffer, the thermal conductivity of the buffer is a very important parameter in the entire disposal safety. For this reason, a lot of research has been conducted on thermal conductivity prediction models that consider various factors. In this study, the thermal conductivity of a buffer is estimated using the machine learning methods of: linear regression, decision tree, support vector machine (SVM), ensemble, Gaussian process regression (GPR), neural network, deep belief network, and genetic programming. In the results, the machine learning methods such as ensemble, genetic programming, SVM with cubic parameter, and GPR showed better performance compared with the regression model, with the ensemble with XGBoost and Gaussian process regression models showing best performance.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.38 no.6
• /
• pp.56-64
• /
• 2001

In this paper, we proposed the simple and selective method for the order of model that reflected the feature of the heart rate variability without the complicated calculation in the power spectral analysis of heart rate variability using autoregressive model. The power spectral analysis of short-term of heart rate variability using autoregressive have been problem to resolution of spectral estimates by the selective model order. As a result that the proposed method for the order comparative tested with the AIC and the fixed order method, the calculation process could become very simple and select the order which correspond with the feature of the time series. We verified it could removed the noisy power components by the fixed order.

• The Journal of the Acoustical Society of Korea
• /
• v.33 no.2
• /
• pp.146-151
• /
• 2014

This paper analyzes the previous Anjok model of the Gayageum and describes a method to improve the frequency modeling based on previous model. In the previous work, relation between the fundamental frequency and Anjok's location on the body is assumed as an exponential function and these frequencies are integrated by a first-order leaky integrator. Finally, a parameter of the formula to calculate the fundamental frequency is obtained by applying integrated frequencies to the linear regression. This model shows 2.5 Hz absolute deviation on average and has maximum error 7.75 Hz for the low fundamental frequencies. In order to overcome this problem, this paper proposes that the Anjok's locations are grouped according to the rate of error increase and linear regression is applied to each group. To find the optimal parameter, the RMSE(Root Mean Square Error) between measured and calculated fundamental frequencies is used. The proposed model shows substantial reduction in errors, especially maximum three times.

• 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.

• Journal of Bio-Environment Control
• /
• v.32 no.2
• /
• pp.115-121
• /
• 2023

The leopard plant has the characteristic of being used for ornamental purposes when there are yellow spots on the leaves, and is widely used as a bed plant for viewing flowers. To set several indicators to predict the growth of crops with ornamental value, and to quantitatively express the relationship between the indicators are necessary. In this study, we determine a model that estimates the leaf area and the number of flower of Farfugium japonicum Kitam. using leaf length and width, and conducting a regression analysis on some regression models. As an indicator for estimating the leaf area and the number of flower, the leaf length and width of F. japonicum were measured and applied to 8 regression models. As a result of regression analysis of 8 models that estimated leaf area and the number of flower, R² values of the linear models were all higher than 0.84 and 0.80. As a result of validation, using the most reliable model among the models for estimating the leaf area and the number of flowering, R² was 0.90 and 0.82, respectively. Using a model that estimates various indicators that can be used for quality evaluation from easy-to-measure morphological factors, the evaluation of ornamental plants will be facilitated.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.33 no.1
• /
• pp.26-32
• /
• 2005

In this paper, the effect of various regression models is investigated on structural optimization using the central composite method. Three bar truss and the upper platform of a satellite are optimized using various regression models that are polynomial, exponential and log functions. Response surface method is non-gradient, semi-global, discrete and fast converging in optimization problem. Sampling points are extracted by the design of experiments using the central composite method. Response surface is generated using the various regression functions. Structural analysis for calculating constraints is executed to find static and dynamic responses. From this study, it is verified that the response surface method has advantage in optimum value and computation time in comparison to other optimization methods.