• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.12
• /
• pp.1772-1781
• /
• 2017

In this study, the design methodology for alleviating the overfitting problem of Polynomial Neural Networks(PNN) is realized with the aid of two kinds techniques such as L2 regularization and Sum of Squared Coefficients (SSC). The PNN is widely used as a kind of mathematical modeling methods such as the identification of linear system by input/output data and the regression analysis modeling method for prediction problem. PNN is an algorithm that obtains preferred network structure by generating consecutive layers as well as nodes by using a multivariate polynomial subexpression. It has much fewer nodes and more flexible adaptability than existing neural network algorithms. However, such algorithms lead to overfitting problems due to noise sensitivity as well as excessive trainning while generation of successive network layers. To alleviate such overfitting problem and also effectively design its ensuing deep network structure, two techniques are introduced. That is we use the two techniques of both SSC(Sum of Squared Coefficients) and $L_2$ regularization for consecutive generation of each layer's nodes as well as each layer in order to construct the deep PNN structure. The technique of $L_2$ regularization is used for the minimum coefficient estimation by adding penalty term to cost function. $L_2$ regularization is a kind of representative methods of reducing the influence of noise by flattening the solution space and also lessening coefficient size. The technique for the SSC is implemented for the minimization of Sum of Squared Coefficients of polynomial instead of using the square of errors. In the sequel, the overfitting problem of the deep PNN structure is stabilized by the proposed method. This study leads to the possibility of deep network structure design as well as big data processing and also the superiority of the network performance through experiments is shown.

• Genomics & Informatics
• /
• v.19 no.1
• /
• pp.11.1-11.8
• /
• 2021

For the novel coronavirus disease 2019 (COVID-19), predictive modeling, in the literature, uses broadly susceptible exposed infected recoverd (SEIR)/SIR, agent-based, curve-fitting models. Governments and legislative bodies rely on insights from prediction models to suggest new policies and to assess the effectiveness of enforced policies. Therefore, access to accurate outbreak prediction models is essential to obtain insights into the likely spread and consequences of infectious diseases. The objective of this study is to predict the future COVID-19 situation of Korea. Here, we employed 5 models for this analysis; SEIR, local linear regression (LLR), negative binomial (NB) regression, segment Poisson, deep-learning based long short-term memory models (LSTM) and tree based gradient boosting machine (GBM). After prediction, model performance comparison was evelauated using relative mean squared errors (RMSE) for two sets of train (January 20, 2020-December 31, 2020 and January 20, 2020-January 31, 2021) and testing data (January 1, 2021-February 28, 2021 and February 1, 2021-February 28, 2021) . Except for segmented Poisson model, the other models predicted a decline in the daily confirmed cases in the country for the coming future. RMSE values' comparison showed that LLR, GBM, SEIR, NB, and LSTM respectively, performed well in the forecasting of the pandemic situation of the country. A good understanding of the epidemic dynamics would greatly enhance the control and prevention of COVID-19 and other infectious diseases. Therefore, with increasing daily confirmed cases since this year, these results could help in the pandemic response by informing decisions about planning, resource allocation, and decision concerning social distancing policies.

• Korean Journal of Environmental Agriculture
• /
• v.38 no.3
• /
• pp.133-138
• /
• 2019

BACKGROUND: Production function gives the equation that shows the relationship between the quantities of productive factors used and the amount of product obtained, and can answer a variety of questions. This study was carried out to evaluate the relationship between irrigation water used for rice production and rice productivity by the production function which shows the mathematical relation between input and output. METHODS AND RESULTS: The statistical data on rice production and on the amount of irrigation water were used for the production function analysis. The analysis period was separated for 1966-1981 and 1982-2011, based on goal's change on agriculture from 'increasing food' to 'complex farming'. The relation between irrigation and yield considering production function is a short-term production function both before and after 1982. These results can be expressed by the sigmoid relation. When comparing the graphs of the two analyzed periods, there are differences in quantity between the maximum point and the minimum point during the same analysis period, which can be called an 'Irrigation Effect' by the difference of irrigation, and 'Technical Effect' by the difference by inputs like as fertilizers etc. CONCLUSION: The results could be useful as information for assessing the relationship between agricultural water and the productivity of rice and predicting rice productivity by irrigation water in Korea.

• Journal of Korean Society on Water Environment
• /
• v.35 no.3
• /
• pp.234-247
• /
• 2019

In a deep lake and reservoir, thermal stratification is of great importance for characteristics of hydrodynamic mixing of the waterbody, and thereby influencesvertical distribution of dissolved oxygen, substances, nutrients, and the phytoplankton community. The purpose of this study, was to project the effect of a future climate change scenario on water temperature, stratification strength, and thermal stability in the Soyanggang Reservoir in the Han River basin of South Korea, using a suite of mathematical models; SWAT, HEC-ResSim, and CE-QUAL-W2(W2). W2 was calibrated with historical data observed 2005-2015. Using climate data generated by HadGEM2-AO with the RCP 4.5 scenario, SWAT predicted daily reservoir inflow 2016-2070, and HEC-ResSim simulated changes in reservoir discharge and water level, based on inflow and reservoir operation rules. Then, W2 was applied, to predict long-term continuous changes of water temperature, in the reservoir. As a result, the upper layer (5 m below water surface) and lower layer (5 m above bottom) water temperatures, were projected to rise $0.0191^{\circ}C/year$(p<0.05) and $0.008^{\circ}C/year$(p<0.05), respectively, in response to projected atmospheric temperature rise rate of $0.0279^{\circ}C/year$(p<0.05). Additionally, with increase of future temperature, stratification strength of the reservoir is projected to be stronger, and the number of the days when temperature difference of the upper layer and the lower layer becomes greater than $5^{\circ}C$, also increase. Increase of water temperature on the surface of the reservoir, affected seasonal growth rate of the algae community. In particular, the growth rate of cyanobacteria increased in spring, and early summer.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.2
• /
• pp.113-118
• /
• 2021

The guided weapon system should ensure economical operation and user safety. In particular, in the case of guided weapon systems developed in the form of a guaranteed bomb, the standards for maintaining reliability considering the long-term storage environment are presented during the development stage, and an optimized inspection cycle is required to maintain this. This study calculated the reliability through a trend test, fitness test, and distribution analysis using a mathematical model based on the maintenance status and shooting results during the inspection period for OO missiles currently in operation for a long time in the military. Through this, it was applied to the inspection period model (Martinez) set during the development stage to determine if the improved inspection period can be utilized. Finally, by synthesizing the data from these studies, a policy management plan was developed according to the extension of the inspection period. The One-Shot system was operated at the inspection period set when it was developed. The study analyzed the actual failure and maintenance data to reset the efficient inspection period.

• Korean Journal of Chemical Engineering
• /
• v.35 no.11
• /
• pp.2241-2255
• /
• 2018

RSM methodology was applied to present mathematical models for the fabrication of polyvinylidene fluoride (PVDF) dual-layer hollow fibers in membrane distillation process. The design of experiments was used to investigate three main parameters in terms of polymer concentration in both outer and inner layers and the flow rate of dope solutions by the Box-Behnken method. According to obtained results, the optimization was done to present the proper membrane with desirable properties. The characteristics of the optimized membrane (named HF-O) suggested by the Box-Behnken (at the predicted point) showed that the proposed models are strongly valid. Then, a morphology study was done to modify the fiber by a combination of three types of a structure such as macro-void, sponge-like and sharp finger-like. It also improved the hydrophobicity of outer surface from 87 to $113^{\circ}$ and the mean pore size of the inner surface from 108.12 to 560.14 nm. The DCMD flux of modified fiber (named HF-M) enhanced 62% more than HF-O when it was fabricated by considering both of RSM and morphology study results. Finally, HF-M was conducted for long-term desalination process up to 100 hr and showed stable flux and wetting resistance during the test. These stepwise approaches are proposed to easily predict the main properties of PVDF dual-layer hollow fibers by valid models and to effectively modify its structure.

• International Journal of Computer Science & Network Security
• /
• v.22 no.11
• /
• pp.11-18
• /
• 2022

Linking Ukraine's education system with the trends of global digitalization is mandatory to ensure the sustainable, long-term development of the country, as well as to increase the sustainability of the education system and the economy as a whole during the crisis period. Now the main problems of the education system in Ukraine are manifested in a complex context caused by Russian armed aggression. In the context of war, problems include differences in adaptation to online learning among educational institutions, limited access to education for vulnerable groups in the zone of active hostilities, the lack of digital educational resources suitable for online learning, and the lack of basic digital skills and competencies among students and teachers necessary to properly conduct online classes. Some of the problems of online learning were solved in the pandemic, but in the context of war Ukrainian society needs a new vision of education and continuous efforts of all social structures in the public and private environment. In the context of war, concerted action is needed to keep education on track and restore it in active zones, adapting to the needs of a dynamic society and an increasingly digitized economy. Among the urgent needs of the education system are a change in the teaching-learning paradigm, which is based on content presentation, memorization, and reproduction, and the adoption of a new, hybrid educational model that will encourage the development of necessary skills and abilities for students and learners in a digitized society and enable citizens close to war zones to learn.

• Journal of agricultural medicine and community health
• /
• v.47 no.2
• /
• pp.61-66
• /
• 2022

Objectives: In the epidemiology of communicable diseases, the term epidemic period, also referred to as "wave" is often used in the general and academic milieu. A wave refers to a natural pattern of increase in the number of sick individuals, a defined peak, and then a decline in the number of cases. It implies a pattern of peaks and valleys after a particular peak is taken. The idea of epidemic waves is a useful tool for predicting the course as well as helping to accurately describe an epidemic. However, in many domestic and foreign news as well as in various research results in Korea, most of the reports either had no standard, were inaccurate, had a questionable classification of the period of the epidemic, or the basis for classification of a given wave was not presented. Methods: The author reviewed and organized related literature with epidemic wave. The author made several suggestions of an epidemic wave as follows. Results: To start with, it should be based on the number of incident cases in consideration of the size of the outbreak, then the period from the bottom to the peak and then reaching the next bottom; also, the period over a certain scale based on the number of incident cases; and the period according to the change in the major infection type (mutation-dominant species). In addition, according to the period of change in the vaccination rate (formation of herd immunity), as well as the content and duration of the intervention, that is, classification according to the applied quarantine stage. Furthermore, the classification of epidemic periods by the time-dependent reproduction number or time-varying reproduction number (Rt), and lastly the application of mathematical methodology. Conclusions: Therefore, classifying the epidemic period into generally known and accepted time frames is considered to be a very important task for future research analysis and development of intervention strategies.

• Journal of Technology Innovation
• /
• v.23 no.4
• /
• pp.33-61
• /
• 2015

The global economy is rapidly changing by technological innovation and diffusion of knowledge across nations. Therefore it is still important issue to find a major variables for convergence and divergence of economic development. The studies up to present on the relationship between innovation and institution has limitations that they have dealt with this issue only in term of cross-sectional study or mathematical research models. This paper aims at analyzing the impact of innovation capabilities and institutional quality on the economic growth. Empirically this paper will explore the relationship among human capital capacity and FDI, R&D expenditures and innovation capabilities and institutional quality. This paper analyzes 64 countries, which were divided into 4 groups depending on the level of economic development. Based on the data from 1995 to 2011 and by using a panel model, we look at the structural implications of the research questions. According to our analysis, the weight of R&D and the innovation capabilities were identified as important determinants of economic growth, and FDI was significant factor for economic growth in the upper middle group countries. In case of the innovation capabilities of countries, the diffusion and openness of innovation were most meaningful variables for economic growth. Also, institutional quality has a significantly positive impact. However, in the low-level economic group, innovation capabilities and institutions have a negative impact on economic growth. This paper identifies an important policy implications that of national innovation and institutional factors should be properly invested in accordance with the level of a country's economic growth.

• Journal of Educational Research in Mathematics
• /
• v.26 no.2
• /
• pp.205-224
• /
• 2016

Nowadays technological instruments are advanced to measure brain waves called EEG. Also, it is important to find some facts that cause students to have mathematic anxiety (MA) and to provide remedy programs to lessen their MA in order to help students cure MA that could contribute to negative self-efficacy toward mathematics and mathematical learning. To find how they change the MA level, a small group of 11 high school students in Suwon city participated for ten weeks at the remedy program based on students' levels of MA diagnosed by MASS instrument (Ko, & Yi, 2011) and proofread by 8 advisors who worked in related research areas. The results showed that the remedy program was effective to lessen students' MA and it should provide a long term period since some negative experiences were accumulated for a long time of his or her past schooling by others such as teachers, peers, and parents. EEG showed that students got better scores on a percent of correct answers and a reaction time and some student' EEG from a group HMA became smaller heights and width in comparison of the other groups.