• Journal of the Korea Convergence Society
• /
• v.10 no.4
• /
• pp.1-6
• /
• 2019

Neural networks have been used to predict the direction of stock index movement from past data. The conventional research that predicts the upward or downward movement of the stock index predicts a rise or fall even with small changes in the index. It is highly likely that losses will occur when trading ETFs by use of the prediction. In this paper, a neural network model that predicts the movement direction of the daily KOrea composite Stock Price Index (KOSPI) to reduce ETF trading losses and earn more than a certain amount per trading is presented. The proposed model has outputs that represent rising (change rate in index ${\geq}{\alpha}$), falling (change rate ${\leq}-{\alpha}$) and neutral ($-{\alpha}$ change rate < ${\alpha}$). If the forecast is rising, buy the Leveraged Exchange Traded Fund (ETF); if it is falling, buy the inverse ETF. The hit ratio (HR) of PNN1 implemented in this paper is 0.720 and 0.616 in the learning and the evaluation respectively. ETF trading yields a yield of 8.386 to 16.324 %. The proposed models show the better ETF trading success rate and yield than the neural network models predicting KOSPI.

• The Korean Journal of Applied Statistics
• /
• v.35 no.4
• /
• pp.543-552
• /
• 2022

When frost occurs, crops are directly damaged. When crops come into contact with low temperatures, tissues freeze, which hardens and destroys the cell membranes or chloroplasts, or dry cells to death. In July 2020, a sudden sub-zero weather and frost hit the Minas Gerais state of Brazil, the world's largest coffee producer, damaging about 30% of local coffee trees. As a result, coffee prices have risen significantly due to the damage, and farmers with severe damage can produce coffee only after three years for crops to recover, which is expected to cause long-term damage. In this paper, we tried to predict frost using frost generation data and weather observation data provided by the Korea Meteorological Administration to prevent severe frost. A model was constructed by reflecting weather factors such as wind speed, temperature, humidity, precipitation, and cloudiness. Using XGB(eXtreme Gradient Boosting), SVM(Support Vector Machine), Random Forest, and MLP(Multi Layer perceptron) models, various hyper parameters were applied as training data to select the best model for each model. Finally, the results were evaluated as accuracy(acc) and CSI(Critical Success Index) in test data. XGB was the best model compared to other models with 90.4% ac and 64.4% CSI, followed by SVM with 89.7% ac and 61.2% CSI. Random Forest and MLP showed similar performance with about 89% ac and about 60% CSI.

• Korean Journal of Environmental Biology
• /
• v.40 no.4
• /
• pp.525-549
• /
• 2022

The process of biological invasion is led by the dynamics of a population as a demographic and evolutionary unit. Spatial structure can affect the population dynamics, and it is worth being considered in research on biological invasion which is always accompanied by dispersal. Metapopulation theory is a representative approach to spatially structured populations, which is chiefly applied in the field of ecology and evolutionary biology despite the controversy about its definition. In this study, metapopulation was considered as a spatially structured population that includes at least one subpopulation with significant extinction probability. The early phase of the invasion is suitable to be analyzed in aspects of the metapopulation concept because the introduced population usually has a high extinction probability, and their ecological·genetic traits determining the invasiveness can be affected by the metapopulation structure. Although it is important in the explanation of the prediction of the invasion probability, the metapopulation concept is rarely used in ecological research about biological invasion in Korea. It is expected that applying the metapopulation theory can supply a more detailed investigation of the invasion process at the population level, which is relatively inadequate in Korea. In this study, a framework dividing the invasive metapopulation into long- and middle-distance scales by the relative distance of movement to the natural dispersal range of species is proposed to easily analyze the effect of a metapopulation in real cases. Increased understanding of the mechanisms underlying invasions and improved prediction of future invasion risk are expected with the metapopulation concept and this framework.

• Korean Journal of Clinical Pharmacy
• /
• v.18 no.2
• /
• pp.84-96
• /
• 2008

In the recent, pharmacokinetic (PK)/pharmacodynamic (PD) modeling has appeared as a critical path tools in new drug development to optimize drug efficacy and safety. PK/PD modeling is the mathematical approaches of the relationships between PK and PD. This approach in new drug development can be estimated inaccessible PK and PD parameters, evaluated competing hypothesis, and predicted the response under new conditions. Additionally, PK/PD modeling provides the information about systemic conditions for understanding the pharmacology and biology. These advantages of PK/PD model development are to provide the early decision-making information in new drug development process, and to improve the prediction power for the success of clinical trials. The purpose of this review article is to summarize the PK/PD modeling process, and to provide the theoretical and practical information about widely used PK/PD models. This review also provides model schemes and the differential equations for the development of PK/PD model.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.19 no.4
• /
• pp.397-402
• /
• 2013

In this study, numerical hull form development of a platform supply vessel, a full scale with the overall length of 26.75m, was performed to predict a bare-hull resistance and a large scale of model tests with a 1/10 scaled model were conducted to verify the success of numerical results. Numerical analysis on heave and pitch motion as a function of encounter frequency and ship's speed for the prediction of seakeeping characteristics are also presented. The experiment results of resistance agreed well with numerical analysis. As a result in the motion response characteristics, the heave RAO indicates high values with the range of encounter frequency 1.8~2.0. The Pitch RAO indicates high motion response characteristics at Beaufort scale No. 3 and 4 in rough seas.

• Journal of The Korean Association of Information Education
• /
• v.18 no.4
• /
• pp.559-566
• /
• 2014

This is to understand the neurological dynamic cognitive processes of math learning based on the abstract mappings( level A2), abstract systems(level A3), and single principles(level A4), which are principles of Fischer's cognitive development theory. Math learning requires flexibility to adapt existing brain function in selecting new neurophysiological activities to learn desired knowledge. This study suggests a general statistical framework for the identification of neurological patterns in different abstract learning change with optimal support. We expected that functional brain networks derived from a simple math learning would change dynamically during the supportive learning associated with different abstract levels. Task based patterns of the brain structure and function on representations of underlying connectivity suggests the possible prediction for the success of the supportive learning.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.20 no.2
• /
• pp.225-230
• /
• 2017

To generate mechanical movements in one-shot devices such as missiles and space launch vehicles, pyrotechnic mechanical device(PMD) such as pin pullers using pyrotechnic charge has been widely used. Reliability prediction of pin pullers is crucial to successfully execute target missions for the one-shot devices. Because the pin pullers require destructive tests to evaluate their reliability, one would need about 3,000 samples of success to guarantee a reliability of 99.9 % with a confidence level of 95 %. This paper suggests the application of a probit model using the charge amount as a functional parameter for estimation of functional reliability of pin puller. To guarantee target reliability, we propose estimation methods of the lower bound of functional reliability by applying the probit model. Given lower bound of functional reliability, we quantitatively show that the optimum amount of charge increases as the number of samples decreases. Along with a variety of simulations the validity of our new model via real test results is confirmed.

• Journal of Gifted/Talented Education
• /
• v.23 no.5
• /
• pp.793-815
• /
• 2013

Research on entrepreneurial gifted youth who have immense potential within the national economy in the near future is critical. The purpose of this article is to establish the proper goal and direction of longitudinal study on entrepreneurial gifted youth and to identify substantial research problems. In order to do a theoretical inspection around preceeding research related with the field, longitudinal study of entrepreneurial youth (including gifted education and entrepreneurship) was executed. As a result, the goal, 'the verification of whether entrepreneurs can be cultivated by educational interventions' was derived as an appropriate research direction. Also, 4 major research problems are presented: (1) Description of the features of entrepreneurial gifted youth and their developmental pattern, (2) Explanation of the effects of educational intervention on intensifying entrepreneurial giftedness, (3) Prediction which identifies the factors that influence the youths' success as eminent entrepreneurs, and (4) Control over promoting talented individuals and refining relevant policy and institutions. Implications and further research directions are also discussed.

• The Journal of the Korean dental association
• /
• v.57 no.12
• /
• pp.736-746
• /
• 2019

Supportive periodontal therapy(SPT) is essential for the long-term success of periodontal treatment. A patient's compliance with SPT is one of the most important factors affecting periodontal status. There are few studies quantifying compliance with SPT. The aim of this study is to quantify patient's compliance using new method and evaluate tooth loss depending on patient's supportive periodontal treatment compliance index(SPTCI) with SPT. This study included subjects diagnosed with generalized moderate to severe chronic periodontitis, who had completed active periodontal treatment and had SPT over 5 years in Wonkwang university dental hospital. Chart review and radiography analysis were performed. To quantify compliance, SPTCI representing average of gap between recommended schedules and actual visits has been used and evaluated with tooth loss. Mean period of SPT was 8.9 years and mean SPTCI was about 120. In statical analysis, patients who have higher SPTCI with SPT are more likely to have higher rate of tooth loss. Under SPTCI of 120, there were no significant co-relation between SPTCI and tooth loss. Patients diagnosed with moderate chronic periodontitis have significant co-relation between SPTCI and tooth loss, whereas patients diagnosed with severe chronic periodontitis have no co-relation. SPTCI, new method of quantifying compliance in this study, affected to tooth loss. This study suggests that using SPTCI could be helpful for prediction of tooth loss and be used to determine the interval of visit.

• KIISE Transactions on Computing Practices
• /
• v.22 no.12
• /
• pp.681-686
• /
• 2016

Software defect prediction is an important factor in efficient project management and success. The severity of the defect usually determines the degree to which the project is affected. However, existing studies focus only on the presence or absence of a defect and not the severity of defect. In this study, we proposed an ensemble model using FCM based on defect severity. The severity of the defect of NASA data set's PC4 was reclassified. To select the input column that affected the severity of the defect, we extracted the important defect factor of the data set using Random Forest (RF). We evaluated the performance of the model by changing the parameters in the 10-fold cross-validation. The evaluation results were as follows. First, defect severities were reclassified from 58, 40, 80 to 30, 20, 128. Second, BRANCH_COUNT was an important input column for the degree of severity in terms of accuracy and node impurities. Third, smaller tree number led to more variables for good performance.