• East Asian mathematical journal
• /
• v.37 no.3
• /
• pp.333-354
• /
• 2021

A support vector machine (SVM) is a state-of-the-art machine learning model rooted in structural risk minimization. SVM is underestimated with regards to its application to real world problems because of the difficulties associated with its use. We aim at showing that the performance of SVM highly depends on which kernel function to use. To achieve these, after providing a summary of support vector machines and kernel function, we constructed experiments with various benchmark datasets to compare the performance of various kernel functions. For evaluating the performance of SVM, the F1-score and its Standard Deviation with 10-cross validation was used. Furthermore, we used taylor diagrams to reveal the difference between kernels. Finally, we provided Python codes for all our experiments to enable re-implementation of the experiments.

• Journal of Wetlands Research
• /
• v.22 no.4
• /
• pp.302-311
• /
• 2020

There are limitations to cope with flood damage by structural strategies alone because both frequency and intensity of floods are increasing due to climate change. Therefore, it is one of the necessary factors in the nonstructural countermeasures to collect and analyze historical flood damage records for the future flood damage assessments. In order to estimate flood damage costs in Gyeongsangbuk-do where severe flood damage occurs frequently due to geographical and climatic effects, this paper has performed the regression analysis on flood damage records over the past 20 years (1999-2018) by rainfall characteristics, which is one of the major causes of flood damage. This paper has then examined the relationship between the terrain features and rainfall characteristics in the regional regression functions, and also estimated the flood damage risk for 100-year rainfall by using the regional regression functions presented for the 22 administrative districts in Gyeongsangbuk-do excluding Ulleung-gun. The flood damage assessment shows that the relatively high damage risk is estimated for county areas adjacent to the eastern coast in Gyeongsangbuk-do. The regional damage estimate functions in this paper are expected to be used as one of the nonstructural countermeasures to estimate flood damage risk for the design or forecasting rainfall data.

• Journal of Korea Water Resources Association
• /
• v.46 no.4
• /
• pp.425-437
• /
• 2013

This study performed the bivariate drought frequency analysis for duration and severity of drought, using copula functions which allow considering the correlation structure of joint features of drought. We suggested the confidence intervals of duration-severity-frequency (DSF) curves for the given drought duration using stochastic scheme of monthly rainfall generation for 57 sites in Korea. This study also investigated drought risk via illustrating the largest drought events on record over 50 and 100 consecutive years. It appears that drought risks are much higher in some parts of the Nakdong River basin, southern and east coastal areas. However, such analyses are not always reliable, especially when the frequency analysis is performed based on the data observed over relatively short period of time. To quantify the uncertainty of drought frequency curves, the droughts were filtered by different durations. The 5%, 25%, 50%, 75%, and 95% confidence intervals of the drought severity for a given duration were estimated based on the simulated rainfall time series. Finally, it is shown that the growing uncertainties is revealed in the estimation of the joint probability using the two marginal distributions since the correlation coefficient of two variables is relatively low.

• Biomedical Science Letters
• /
• v.17 no.4
• /
• pp.321-328
• /
• 2011

This study divided a group of healthy adults aged 20 or older who had a health examination at J General Hospital in Gyeonggi Province into three groups according to the degrees of metabolic syndrome risk factors. They include the normal group (n=58), the pre-metabolic syndrome group (n=112) and the metabolic syndrome group (n=32). They were compared in exercise capacity and cardiac structure and function and impacts of exercise capacity on the cardiac diastolic function. All the groups took echocardiography to have their cardiac structures and functions examined and an exercise stress test to have their exercise capacity measured. The research findings were as follows: There were differences in exercise capacity, cardiac structure, and diastolic heart function among three groups. Between exercise capacity and diastolic heart function was found to be related. It turned out exercise capacity affected the cardiac diastolic functions. In conclusion, there were significant differences in exercise capacity between the normal group and the metabolic syndrome group and in the cardiac structure and function among the normal, metabolic syndrome, and pre-metabolic syndrome group. In addition, METs (metabolic equivalents) and heart rate recovery of exercise capacity turned out to affect cardiac diastolic functions.

• Biomedical Science Letters
• /
• v.22 no.3
• /
• pp.107-114
• /
• 2016

Maintenance of fasting blood glucose levels is important for glucose homeostasis. Disruption of feedback mechanisms are a major reason for elevations of glucose level in blood, which is a risk factor for type 2 diabetes mellitus that is mainly caused by malfunction of pancreatic beta-cell and insulin. The fasting blood glucose level has been known to be influenced by genetic and environmental factors. Mitochondria have many functions for cell survival and death: glucose metabolism, fatty acid oxidation, ATP generation, reactive oxygen species (ROS) metabolism, calcium handling, and apoptosis regulation. In addition to these functions, mitochondria change their morphology dynamically in response to multiple signals resulting in fusion and fission. In this study, we aimed to examine association between fasting blood glucose levels and variants of the genes that are reported to have functions in mitochondrial dynamics, fusion and fission, using a cohort study. A total 416 SNPs from 36 mitochondrial dynamics genes were selected to analyze the quantitative association with fasting glucose level. Among the 416 SNPs, 4 SNPs of PRKACB, 13 SNPs of PPP3CA, 6 SNPs of PARK2, and 3 SNPs of GDAP1 were significantly associated. In this study, we were able to confirm an association of mitochondrial dynamics genes with glucose levels. To our knowledge our study is the first to identify specific SNPs related to fasting blood glucose level.

• Proceedings of the KSR Conference
• /
• 2011.10a
• /
• pp.2473-2478
• /
• 2011

The asset management of railway facilities is a total framework for finally supporting a safe and comfortable train service, which includes functions of supporting evaluation of condition and performance of infrastructures, making the decision method of repair or rehabilitation of deteriorated facilities, and lengthening the life cycle of structure through the decision of adequate cost and time of repair or reinforcement. In the range of the asset management, organization, human, the target, and information & data of company are included. Therefore, in this paper, appling the method of asset management analysis to the railway structures, the process of the risk assesment using BRE(Business Risk Exposure) and the basis of consisting optimized renewal decision-making are expressed.

• International Journal of Control, Automation, and Systems
• /
• v.2 no.1
• /
• pp.39-54
• /
• 2004

We developed an automatic blood pressure control system to maintain the blood pressure of patients at a substantially low level during a surgical operation. The developed system discharges two functions, continuous feedback control of the mean arterial pressure (MAP) by a state-predictive servo controller and risk control based on the inference by fuzzy-like logics and rules using measured data. Twenty-eight clinical applications were made beginning in November 1995, and the effects of the automatic blood pressure control on the operation time and on bleeding were assessed affirmatively by means of Wilcoxon testing. This paper essentially reports the engineering details of the control system.

• IE interfaces
• /
• v.15 no.3
• /
• pp.270-278
• /
• 2002

Product Liability(PL) is a legal policy to deal with global competition by improving domestic industrial competitive power and to reduce the cost of defect products. The purpose of this paper is to address the state of the art solutions to dispute on PL, in reality of a frequent occurrence of global product exchange focussing on product safety that is one of the most important functions of PL and to improve solution of the product safety and reliability responsive to PL. To minimize PL exposure, manufacturers should reflect comprehensive product safety and reliability concepts in establishing PL prevention policies. Total PL prevention policies are composed of total quality management and product safety management system in respect of safety design, risk, and reliability. These PL prevention activities should be performed consistently during the total product life cycle, especially product research and development periods.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.14 no.4
• /
• pp.249-273
• /
• 2010

Often in practice, the implied volatility of an option is calculated to find the option price tomorrow or the prices of, nearby' options. To show that one does not need to adhere to the Black- Scholes formula in this scheme, Figlewski has provided a new pricing formula and has shown that his, alternating passive model' performs as well as the Black-Scholes formula [8]. The Figlewski model was modified by Henderson et al. so that the formula would have no static arbitrage [10]. In this paper, we show how to construct a huge class of such static no arbitrage pricing functions, making use of distortions, coherent risk measures and the pricing theory in incomplete markets by Carr et al. [4]. Through this construction, we provide a more elaborate static no arbitrage pricing formula than Black-Sholes in the above scheme. Moreover, using our pricing formula, we find a volatility curve which fits with striking accuracy the synthetic data used by Henderson et al. [10].

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.19 no.4
• /
• pp.429-441
• /
• 2015

We study optimal consumption/investment and life insurance purchase rules for a wage earner with mortality risk under regime-switching financial market conditions, in a continuous time-horizon. We apply the Markov chain approximation method and suggest an efficient algorithm using parallel computing to solve the simultaneous Hamilton-Jaccobi-Bellman equations arising from the optimization problem. We provide numerical results under the utility functions of the constant relative risk aversion type, with which we illustrate the effects of regime switching on the optimal policies by comparing them with those in the absence of regime switching.