• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.56 no.2
• /
• pp.243-252
• /
• 2023

Fluctuations in the price of aquaculture products have recently intensified. In particular, wholesale price fluctuations are adversely affecting consumers. Therefore, there is an emerging need for a study on forecasting the wholesale price of aquaculture products. The present study forecasted the wholesale price of olive flounder Paralichthys olivaceus, a representative farmed fish species in Korea, by constructing multivariate long-short term memory (LSTM) and gated recurrent unit (GRU) models. These deep learning models have recently been proven to be effective for forecasting in various fields. A total of 191 monthly data obtained for 17 variables were used to train and test the models. The results showed that the mean average percent error of LSTM and GRU models were 2.19% and 2.68%, respectively.

• Journal of Korean Society for Atmospheric Environment
• /
• v.29 no.3
• /
• pp.297-306
• /
• 2013

This study used long-term air and weather data from 2000 to 2009 as raw data sets to develop regression models in order to estimate precipitation scavenging contributions of ambient $PM_{10}$ and $NO_2$ in Korea. The data were initially analyzed to calculate scavenging ratio (SR), defined as the removal efficiency for $PM_{10}$ and $NO_2$ by actual precipitation. Next, the effective scavenging contributions (ESC) with considering precipitation probability density were calculated for each sector of precipitation range. Finally, the empirical regression equations for the two air pollutants were separately developed, and then the equations were applied to test the model validity with the raw data sets of 2010 and 2011, which were not involved in the modeling process. The results showed that the predicted $PM_{10}$ ESC by the model was 23.8% and the observed $PM_{10}$ ESCs were 23.6% in 2010 and 24.0% in 2011, respectively. As for $NO_2$, the predicted ESC by the model was 16.3% and the observed ESCs were 16.4% in 2010 and 16.6% in 2011, respectively. Thus the developed regression models fitted quite well the actual scavenging contribution for both ambient $PM_{10}$ and $NO_2$. The models can then be used as a good tool to quantitatively apportion the natural and anthropogenic sink contribution in Korea. However, to apply the models for far future, the precipitation probability density function (PPDF) as a weather variable in the model equations must be renewed periodically to increase prediction accuracy and reliability. Further, in order to apply the models in a specific local area, it is recommended that the long-term oriented local PPDF should be inserted in the models.

• Korean Journal of Biological Psychiatry
• /
• v.27 no.1
• /
• pp.18-26
• /
• 2020

Objectives The aim was to find effective vectorization and classification models to predict a psychiatric diagnosis from text-based medical records. Methods Electronic medical records (n = 494) of present illness were collected retrospectively in inpatient admission notes with three diagnoses of major depressive disorder, type 1 bipolar disorder, and schizophrenia. Data were split into 400 training data and 94 independent validation data. Data were vectorized by two different models such as term frequency-inverse document frequency (TF-IDF) and Doc2vec. Machine learning models for classification including stochastic gradient descent, logistic regression, support vector classification, and deep learning (DL) were applied to predict three psychiatric diagnoses. Five-fold cross-validation was used to find an effective model. Metrics such as accuracy, precision, recall, and F1-score were measured for comparison between the models. Results Five-fold cross-validation in training data showed DL model with Doc2vec was the most effective model to predict the diagnosis (accuracy = 0.87, F1-score = 0.87). However, these metrics have been reduced in independent test data set with final working DL models (accuracy = 0.79, F1-score = 0.79), while the model of logistic regression and support vector machine with Doc2vec showed slightly better performance (accuracy = 0.80, F1-score = 0.80) than the DL models with Doc2vec and others with TF-IDF. Conclusions The current results suggest that the vectorization may have more impact on the performance of classification than the machine learning model. However, data set had a number of limitations including small sample size, imbalance among the category, and its generalizability. With this regard, the need for research with multi-sites and large samples is suggested to improve the machine learning models.

• Journal of Korea Soil Environment Society
• /
• v.3 no.1
• /
• pp.31-44
• /
• 1998

The geochemical experiments were carried out to investigate a removal effect of heavy metals in abdndoned metallic mine wastes, and to conceive a treatment techniques of them. In order to prevent contamination, experiment appature was made of acrylic acid resin and polyethylene which resist to acid and alkali. Experiment models are devided into four groups based on the system environments, distribution patterns and a kind of filling materials. The first group is background model(model I ) which is filled with waste only and opened to air. The second one is four layer group which is subdivided into two models, opened and closed systems, and the third mix group which is subdivided into three models based on mixing ratio of filling materials and system environment like a layered group. The forth is composed of two layer model, lower one composed of waste and upper one limestone chips. Solution drained from Model Ishows a high contents of heavy metals on the all terms of experiments. Among the models, however, the closed mix model V and Ⅶ show the most effective removal of heavy metals liberated from wastes. Models having different mixing ratios of filling materials on closed systems does not affect in heavy metal removal effect. But, the distribution patterns of filling materials affect very much on removal effect of heavy metals. The closed models with same constitution ratios and distribution patterns of filling materials show more and less effective removal to the open models.

• Steel and Composite Structures
• /
• v.29 no.4
• /
• pp.465-481
• /
• 2018

The use of external carbon-fiber-reinforced polymer (CFRP) laminates is one of the most effective techniques existing for the confinement of circular concrete specimens. Currently, several researches have been made to develop models for predicting the ultimate conditions of this type of confinement. As most of the major existing models were developed based on limited experimental database. This paper presents the development of new confinement ultimate conditions, strength and strain models, for concrete cylinders confined with CFRP composites based on a statistical analysis of a large existing experimental database of 310 cylindrical concrete specimens wrapped with CFRP. The database is used to evaluate the performance of the proposed and major existing strength and strain models. Based on the two different statistical indices, the coefficient of determination ($R^2$) and the Root Mean Square Error (RMSE), the two proposed confinement ultimate conditions presents a good performance compared to the major existing models except the models of Lam and Teng (2003) and Youssef et al. (2007) which have relatively similar performance to the proposed models.

• Journal of Korean Association for Spatial Structures
• /
• v.13 no.1
• /
• pp.87-96
• /
• 2013

This study investigated characteristics of buckling load and effective buckling length by member rigidity of dome-typed space frame which was sensitive to initial conditions. A critical point and a buckling load were computed by analyzing the eigenvalues and determinants of the tangential stiffness matrix. The hexagonal pyramid model and star dome were selected for the case study in order to examine the nodal buckling and member buckling in accordance with member rigidity. From the numerical results, an effective buckling length factor of adopted models was bigger than that of Euler buckling for the case of fixed boundary. These numerical models indicated that the influence of nodal buckling was greater than that of member buckling as member rigidity was higher. Besides, there was a tendency that the bifurcation appeared on the equilibrium path before limit point in the member buckling model.

• Journal of Korean Institute of Industrial Engineers
• /
• v.37 no.2
• /
• pp.118-123
• /
• 2011

This study aims to investigate stereotype and effective cue presentation methods for burner-control relationship of four-stove range for Korean. A total of 381 subjects(male : 262, female : 119) were surveyed using questionnaire, in which eight gas range models drawn by 3-D Max 2010 were presented. The gas range models were labeled by signs(☆#${\triangle}{\square}$) to eliminate suggestive effect of sequential codes such as alphabets and numbers. The results showed that the stereotype was significantly affected by occupation(p < 0.10), but not by subjects' gender and age(p > 0.39). The stereotype of four-burner gas range for Korean was the same as that of Chinese, while the stereotype was different from that of American. The cues with rectangular-shaped arrangements identical or similar to those of burners were effective to relate burners to corresponding controls. The diamond-shaped cues and burner arrangements were not appropriate for representing burner-control relationship of four-stove gas range. These findings would be used as a basic guideline when designing four-burner gas range or similar equipments.

• Nuclear Engineering and Technology
• /
• v.41 no.7
• /
• pp.929-944
• /
• 2009

The present study is concerned with the extension of the Effective Convectivity Model (ECM) to the phase-change problem to simulate the dynamics of the melt pool formation in a Light Water Reactor (LWR) lower plenum during hypothetical severe accident progression. The ECM uses heat transfer characteristic velocities to describe turbulent natural convection of a melt pool. The simple approach of the ECM method allows implementing different models of the characteristic velocity in a mushy zone for non-eutectic mixtures. The Phase-change ECM (PECM) was examined using three models of the characteristic velocities in a mushy zone and its performance was compared. The PECM was validated using a dual-tier approach, namely validations against existing experimental data (the SIMECO experiment) and validations against results obtained from Computational Fluid Dynamics (CFD) simulations. The results predicted by the PECM implementing the linear dependency of mushy-zone characteristic velocity on fluid fraction are well agreed with the experimental correlation and CFD simulation results. The PECM was applied to simulation of melt pool formation heat transfer in a Pressurized Water Reactor (PWR) and Boiling Water Reactor (BWR) lower plenum. The study suggests that the PECM is an adequate and effective tool to compute the dynamics of core melt pool formation.

• Structural Engineering and Mechanics
• /
• v.67 no.2
• /
• pp.155-163
• /
• 2018

The Early-age construction loading and changing properties of concrete, especially in the multi-story structures can affect the slab deflection, significantly. Based on previously conducted experiment on eight simply-supported one-way slabs this paper investigates the effect of concrete type, fiber type and content, loading value, cracking moment, ultimate moment and applied moment on the instantaneous deflection of Self-Compacting Concrete (SCC) slabs. Two distinct loading levels equal to 30% and 40% of the ultimate capacity of the slab section were applied on the slabs at the age of 14 days. A wide range of the existing models of the effective moment of inertia which are mainly developed for conventional concrete elements, were investigated. Comparison of the experimental deflection values with predictions of the existing models shows considerable differences between the recorded and estimated instantaneous deflection of SCC slabs. Calculated elastic deflection of slabs at the ages of 14 and 28 days were also compared with the experimental deflection of slabs. Based on sensitivity analysis of the effective parameters, a new model is proposed and verified to predict the effective moment of inertia in SCC slabs with and without fiber reinforcing under two different loading levels at the age of 14 days.

• Nuclear Engineering and Technology
• /
• v.49 no.3
• /
• pp.459-465
• /
• 2017

The goal of this study is to evaluate the effective thermal conductivity and diffusivity of containment walls as heat sinks or passive cooling systems during nuclear power plant (NPP) accidents. Containment walls consist of steel reinforced concrete, steel liners, and tendons, and provide the main thermal resistance of the heat sinks, which varies with the volume fraction and geometric alignment of the rebar and tendons, as well as the temperature and chemical composition. The target geometry for the containment walls of this work is the standard Korean NPP OPR1000. Sample tests and numerical simulations are conducted to verify the correlations for models with different densities of concrete, volume fractions, and alignments of steel. Estimation of the effective thermal conductivity and diffusivity of the containment wall models is proposed. The Maxwell model and modified Rayleigh volume fraction model employed in the present work predict the experiment and finite volume method (FVM) results well. The effective thermal conductivity and diffusivity of the containment walls are summarized as functions of density, temperature, and the volume fraction of steel for the analysis of the NPP accidents.