• The Korean Journal of Applied Statistics
• /
• v.30 no.1
• /
• pp.41-55
• /
• 2017

This study predicts the average scores of top 150 PGA golf players on 132 PGA Tour tournaments (2013-2015) using data mining techniques and statistical analysis. This study also aims to predict the Top 10 and Top 25 best players in 4 different playoffs. Linear and nonlinear regression methods were used to predict average scores. Stepwise regression, all best subset, LASSO, ridge regression and principal component regression were used for the linear regression method. Tree, bagging, gradient boosting, neural network, random forests and KNN were used for nonlinear regression method. We found that the average score increases as fairway firmness or green height or average maximum wind speed increases. We also found that the average score decreases as the number of one-putts or scrambling variable or longest driving distance increases. All 11 different models have low prediction error when predicting the average scores of PGA Tournaments in 2015 which is not included in the training set. However, the performances of Bagging and Random Forest models are the best among all models and these two models have the highest prediction accuracy when predicting the Top 10 and Top 25 best players in 4 different playoffs.

• Ecology and Resilient Infrastructure
• /
• v.9 no.1
• /
• pp.24-35
• /
• 2022

For water quality management, it is necessary to continuously improve the forecasting by analyzing the past water quality, and a Data-driven model is emerging as an alternative. Because the Data-driven model is built based on a wide range of data, it is essential to apply the correlation analysis method for the combination of input variables to obtain more reliable results. In this study, the Gamma Test was applied as a preceding step to build a faster and more accurate data-driven water quality prediction model. First, a physical-based model (HSPF, EFDC) was operated to produce daily water quality reflecting the complexity of the watershed according to various hydrological conditions for Paldang Dam. The Gamma Test was performed on the water quality at the water quality prediction site (Paldangdam2) and major rivers flowing into the Paldang Dam, and the method of selecting the optimal input data combination was presented through the analysis results (Gamma, Gradient, Standar Error, V-Ratio). As a result of the study, the selection criteria for a more efficient combination of input data that can save time by omitting trial and error when building a data-driven model are presented.

• Geophysics and Geophysical Exploration
• /
• v.10 no.2
• /
• pp.147-153
• /
• 2007

The conjugate gradient (CG) method is one of the most efficient algorithms for solving a linear system of equations. In addition to being used as a linear equation solver, it can be applied to a least-squares problem. When the CG method is applied to large-scale three-dimensional inversion of magnetotelluric data, two approaches have been pursued; one is the linear CG inversion in which each step of the Gauss-Newton iteration is incompletely solved using a truncated CG technique, and the other is referred to as the nonlinear CG inversion in which CG is directly applied to the minimization of objective functional for a nonlinear inverse problem. In each procedure we only need to compute the effect of the sensitivity matrix or its transpose multiplying an arbitrary vector, significantly reducing the computational requirements needed to do large-scale inversion.

• Journal of the Korean Geographical Society
• /
• v.39 no.5 s.104
• /
• pp.722-734
• /
• 2004

Longitudinal profiles of bedrock rivers play a fundamental role in landscape history by setting the boundary conditions for landform evolution. Longitudinal profiles are changed with climatic conditions, lithology and tectonic movements. Tectonic movement is an important factor controlling longitudinal profiles, especially in tectonically active area where uplift rates are regarded as a major factor controlling channel gradient. However study on bedrock channel has made little progress, because controls over bedrock river incision are yet to be clarified. Previous numerical simulations have used a simple diffusion model, which links together the overall processes of bedrock channel erosion as in other landform evolution models. In this study, previous bedrock incision models based on physical processes (especially abrasion) are reviewed and new modifications are introduced. Using newly formulated numerical model, the role of spatial pattern and intensity of tectonic uplift on changes in river longitudinal profile was simulated and discussed.

• Journal of Korea Water Resources Association
• /
• v.44 no.3
• /
• pp.249-262
• /
• 2011

In this study, in order to reduce the numerical oscillation due to the unbalance between source and flux terms as the HLLC scheme is applied to the flow analysis on the irregular bed topography, a unstructured finite volume model based on the well-balanced HLLC scheme and the shallow water equations is developed and applied to problems of dam-break waves. The well-balanced HLLC scheme considers directly the gradient of bed topography as the flux terms is calculated. This scheme provides the good numerical balance between the source and flux terms in the case of the application to the steady-state transcritical flow. To verify the numerical model developed in this study, it is applied to three cases of hydraulic model experiments and a field case study of Mapasset dam failure (France). As a result of the verification, the predicted numerical results agree relatively well with available laboratory and field measurements. The model provides slightly more accurate results compared with the existing models.

• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.119-133
• /
• 2022

One of the most important current features of food related industry is the growth of food delivery service. Another notable food related culture is, with the advent of Youtube, the popularity of Mukbang, which refers to content that records eating. Based on these background, this study intended to focus on two things. First, we tried to see the impact of Youtube Mukbang and the sentiments of Mukbang comments on the number of related food deliveries. Next, we tried to set up the predictive modeling of chicken delivery order with machine learning method. We used Youtube Mukbang comments data as well as weather related data as main independent variables. The dependent variable used in this study is the number of delivery order of fried chicken. The period of data used in this study is from June 3, 2015 to September 30, 2019, and a total of 1,580 data were used. For the predictive modeling, we used machine learning methods such as linear regression, ridge, lasso, random forest, and gradient boost. We found that the sentiment of Youtube Mukbang and comments have impacts on the number of delivery orders. The prediction model with Mukban data we set up in this study had better performances than the existing models without Mukbang data. We also tried to suggest managerial implications to the food delivery service industry.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.5B
• /
• pp.429-439
• /
• 2009

When Catastrophic extreme flood occurs due to dam break, the response time for flood warning is much shorter than for natural floods. Numerical models can be powerful tools to predict behaviors in flood wave propagation and to provide the information about the flooded area, wave front arrival time and water depth and so on. But flood wave propagation due to dam break can be a process of difficult mathematical characterization since the flood wave includes discontinuous flow and dry bed propagation. Nevertheless, a lot of numerical models using finite volume method have been recently developed to simulate flood inundation due to dam break. As Finite volume methods are based on the integral form of the conservation equations, finite volume model can easily capture discontinuous flows and shock wave. In this study the numerical model using Riemann approximate solvers and finite volume method applied to the conservative form for two-dimensional shallow water equation was developed. The MUSCL scheme with surface gradient method for reconstruction of conservation variables in continuity and momentum equations is used in the predictor-corrector procedure and the scheme is second order accurate both in space and time. The developed finite volume model is applied to 2D partial dam break flows and dam break flows with triangular bump and validated by comparing numerical solution with laboratory measurements data and other researcher's data.

• Journal of Korean Society of Disaster and Security
• /
• v.12 no.2
• /
• pp.73-82
• /
• 2019

Recently, as the occurrence frequency of sudden floods due to climate change increased, the flood damage on riverside social infrastructures was extended so that there has been a threat of overflow. Therefore, a rapid prediction of potential flooding in riverside social infrastructure is necessary for administrators. However, most current flood forecasting models including hydraulic model have limitations which are the high accuracy of numerical results but longer simulation time. To alleviate such limitation, data driven models using artificial neural network have been widely used. However, there is a limitation that the existing models can not consider the time-series parameters. In this study the water surface elevation of the Hangang River bridge was predicted using the NARX model considering the time-series parameter. And the results of the ANN and RNN models are compared with the NARX model to determine the suitability of NARX model. Using the 10-year hydrological data from 2009 to 2018, 70% of the hydrological data were used for learning and 15% was used for testing and evaluation respectively. As a result of predicting the water surface elevation after 3 hours from the Hangang River bridge in 2018, the ANN, RNN and NARX models for RMSE were 0.20 m, 0.11 m, and 0.09 m, respectively, and 0.12 m, 0.06 m, and 0.05 m for MAE, and 1.56 m, 0.55 m and 0.10 m for peak errors respectively. By analyzing the error of the prediction results considering the time-series parameters, the NARX model is most suitable for predicting water surface elevation. This is because the NARX model can learn the trend of the time series data and also can derive the accurate prediction value even in the high water surface elevation prediction by using the hyperbolic tangent and Rectified Linear Unit function as an activation function. However, the NARX model has a limit to generate a vanishing gradient as the sequence length becomes longer. In the future, the accuracy of the water surface elevation prediction will be examined by using the LSTM model.

• Journal of Korea Water Resources Association
• /
• v.45 no.4
• /
• pp.349-360
• /
• 2012

A horizontally curvilinear non-hydrostatic free surface model that was applicable to three-dimensional viscous flows was developed. The proposed model employed a top-layer equation to close kinematic free-surface boundary condition, and an isotropic k-${\varepsilon}$ model to close turbulence viscosity in the Reynolds averaged Navier-Stokes equation. The model solved the governing equations with a fractional step method, which solved intermediate velocities in the advection-diffusion step, and corrects these provisional velocities by accounting for source terms including pressure gradient and gravity acceleration. Numerical applications were implemented to the wind-driven currents in a two-dimensional closed basin, the flow in a steep-sided trench, and the flow in a strongly-curved channel accounting for secondary current by the centrifugal force. Through the numerical simulations, the model showed its capability that were in good agreement with experimental data with respect to free surface elevation, velocity, and turbulence characteristics.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2005.05b
• /
• pp.424-428
• /
• 2005

본 연구에서는 임의로 변화하는 지형상에 적용시에 보존 특성이 성립하는 쌍곡선형 천수방정식 해석 기법을 개발하였다. 일반적으로 쌍곡선형의 천수방정식은 상류와 사류를 쉽고 정확하게 해석할 수 있고, 또한 Euler 방정식 해석기법을 이용한 다양한 해석기법이 개발되어 있다는 장점을 지니고 있다. 그러나 바닥지형이 변화하는 경우, 생성항과 플럭스항 사이에 수치적 해석기법 차이에서 발생하는 수치적 불균형이 발생하여 수치모형의 적용성이 현저하게 저하된다. 따라서 본 연구에서는 이와 같은 현상을 개선하기 위하여, 기존의 표면경사법을 개선한 기법을 제시하였다. MUSCL-Hancock 기법과 HLLC 근사 Riemann 기법을 이용하였으며, 플럭스항과 수치적 균형을 이루기 위한 이산화기법을 제안하였다. 모형의 검증을 위하여 정상류 상태의 상류와 사류 해석을 수행하였고, 마른바닥에서의 댐붕괴파와 수직한 지형 변화를 갖는 수로상의 서지의 진행 등과 같은 부정류에 대하여 적용하였다. 적용결과, 매우 정확하고 수치적으로 안정된 계산결과를 얻었다.