• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.43 no.4
• /
• pp.107-115
• /
• 2020

Support vector regression (SVR) is devised to solve the regression problem by utilizing the excellent predictive power of Support Vector Machine. In particular, the ⲉ-insensitive loss function, which is a loss function often used in SVR, is a function thatdoes not generate penalties if the difference between the actual value and the estimated regression curve is within ⲉ. In most studies, the ⲉ-insensitive loss function is used symmetrically, and it is of interest to determine the value of ⲉ. In SVQR (Support Vector Quantile Regression), the asymmetry of the width of ⲉ and the slope of the penalty was controlled using the parameter p. However, the slope of the penalty is fixed according to the p value that determines the asymmetry of ⲉ. In this study, a new ε-insensitive loss function with p1 and p2 parameters was proposed. A new asymmetric SVR called GSVQR (Generalized Support Vector Quantile Regression) based on the new ε-insensitive loss function can control the asymmetry of the width of ⲉ and the slope of the penalty using the parameters p1 and p2, respectively. Moreover, the figures show that the asymmetry of the width of ⲉ and the slope of the penalty is controlled. Finally, through an experiment on a function, the accuracy of the existing symmetric Soft Margin, asymmetric SVQR, and asymmetric GSVQR was examined, and the characteristics of each were shown through figures.

• Journal of the Korean Data and Information Science Society
• /
• v.15 no.2
• /
• pp.449-455
• /
• 2004

Support vector machine (SVM) has been very successful in pattern recognition and function estimation problems for crisp data. This paper proposes a new method to evaluate interval regression models for crisp input-output data. The proposed method is based on quadratic loss SVM, which implements quadratic programming approach giving more diverse spread coefficients than a linear programming one. The proposed algorithm here is model-free method in the sense that we do not have to assume the underlying model function. Experimental result is then presented which indicate the performance of this algorithm.

• The Journal of the Korea Contents Association
• /
• v.15 no.7
• /
• pp.255-263
• /
• 2015

The current study examines the association between multiple losses and depressive symptoms and the role of social support in multiple losses. Using a prospective designed dataset(Changing Lives of Older Couples), this study found a significant difference on the depressive symptom levels between multiple losses and single loss. In addition, there was a significant buffering effect of social support in bereavement, as oppose to previous literature that social support does not buffer the initial bereavement reaction in comparisons between the bereaved with multiple losses and the bereaved with a single loss. The author discusses the importance of monitoring elderly people with multiple losses and availability of social support before and after the loss.

• Structural Engineering and Mechanics
• /
• v.61 no.4
• /
• pp.527-538
• /
• 2017

The evaluation of the loss-of-support conditions of frictional beam-to-column connections using simplified numerical models describing the transverse response of a portal-like structure is presented in this paper considering the effects of the seismic-hazard disaggregation. Real earthquake time histories selected from European Strong-motion Database (ESD) are used to show the effects of the seismic-hazard disaggregation on the beam loss-of-support conditions. Seismic events are classified according to different values of magnitudes, epicentral distances and soil conditions (stiff or soft soil) highlighting the importance of considering the characteristics of the seismic input in the assessment of the loss-of-support conditions of frictional beam-to-column connections. A rigid and an elastic model of a frame of a precast industrial building (2-DoF portal-like model) are presented and adopted to find the minimum required friction coefficient to avoid sliding. Then, the mean value of the minimum required friction coefficient with an epicentral distance bin of 10 km is calculated and fitted with a linear function depending on the logarithm of the epicentral distance. A complete parametric analysis varying the horizontal and vertical period of vibration of the structure is performed. Results show that the loss-of-support condition is strongly influenced by magnitude, epicentral distance and soil conditions determining the frequency content of the earthquake time histories and the correlation between the maxima of the horizontal and vertical components. Moreover, as expected, dynamic characteristics of the structure have also a strong influence. Finally, the effect of the column nonlinear behavior (i.e. formation of plastic hinges at the base) is analyzed showing that the connection and the column are a series system where the maximum force is limited by the element having the minimum strength. Two different longitudinal reinforcement ratios are analyzed demonstrating that the column strength variation changes the system response.

• Korean Journal of Social Welfare
• /
• v.42
• /
• pp.397-422
• /
• 2000

Since 1998 when economic assistance from IMF started, the empirical research on the effects of unemployment and drastic income loss on psychological functioning among the unemployed as well as their family members has increased. These studies have found that unemployment and drastic income loss following unemployment have significant negative influence on the psychological outcomes such as anxiety and depression of the unemployed and their family members. Studies have also focused on the role of social support in this process, and reported that depending on the levels of received social support, unemployment and income loss have differential effects on the psychological aspects of the unemployed. However, these studies have several weaknesses. First, most of the related studies employed the data which were collected from limited regions of the country, which imposes limitations on the scope of the generalizability of research findings. Second, the main independent variables used in these studies were mainly unemployment or income loss, which ignore the psychological evaluation by the employed of their family financial situations. Third, in analyzing the moderating effects of social support, most studies have focused mainly on showing the existence of moderating effects by social support. Consequently, the nature and role of social support remained unanalyzed and left to speculations. The purpose, of this study is to examine the effects of economic pressure experienced by family heads who were unemployed and to analyze the moderating role of social support based on a nationally representative sample. The findings showed that economic pressure has negative influence on anxiety and depressive feelings among the unemployed, and that the effect of economic pressure on depressive feelings were substantially higher among those who have received lower levels of social support from family members than that among those with higher levels of family support.

• Communications for Statistical Applications and Methods
• /
• v.18 no.2
• /
• pp.165-170
• /
• 2011

Support vector quantile regression(SVQR) is capable of providing a good description of the linear and nonlinear relationships among random variables. In this paper we propose a sparse SVQR to overcome a limitation of SVQR, nonsparsity. The asymmetric e-insensitive loss function is used to efficiently provide sparsity. The experimental results are presented to illustrate the performance of the proposed method by comparing it with nonsparse SVQR.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.4
• /
• pp.749-755
• /
• 2009

Support vector classification (SVC) provides more complete description of the lin-ear and nonlinear relationships between input vectors and classifiers. In this paper. we propose the sparse kernel classifier to solve the optimization problem of classification with a modified hinge loss function and absolute loss function, which provides the efficient computation and the sparsity. We also introduce the generalized cross validation function to select the hyper-parameters which affects the classification performance of the proposed method. Experimental results are then presented which illustrate the performance of the proposed procedure for classification.

• Journal of the Korean Data and Information Science Society
• /
• v.17 no.3
• /
• pp.905-912
• /
• 2006

Support vector classification(SVC) provides more complete description of the linear and nonlinear relationships between input vectors and classifiers. In this paper we propose to solve the optimization problem of SVC with a modified hinge loss function, which enables to use an iterative reweighted least squares(IRWLS) procedure. We also introduce the approximate cross validation function to select the hyperparameters which affect the performance of SVC. Experimental results are then presented which illustrate the performance of the proposed procedure for classification.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.315-315
• /
• 2000

In this paper, we perform the time series prediction using the SVM(Support Vector Machine). We make use of two different loss functions and two different kernel functions; i) Quadratic and $\varepsilon$-insensitive loss function are used; ii) GRBF(Gaussian Radial Basis Function) and ERBF(Exponential Radial Basis Function) are used. Mackey-Glass time series are used for prediction. For both cases, we compare the results by the SVM to those by ANN(Artificial Neural Network) and show the better performance by SVM than that by ANN.

• Journal of Korean Critical Care Nursing
• /
• v.10 no.2
• /
• pp.34-44
• /
• 2017

Purpose: The purpose of this study was to identify post-operative discomfort, sense of loss, family support, and resilience in breast cancer surgery patients, and to investigate factors that affect resilience. Method: The sample of this study consisted of 108 patients who underwent surgery for breast cancer in two university hospitals located in B city. The collected data was analyzed with descriptive statistics, t-test, ANOVA, and Scheffé test, Pearson's correlation coefficients, and hierarchical multiple regression. Results: The factors that significantly affected resilience were as follows: having a religion (${\beta}=-.20$, p=.006), having an occupation (${\beta}=.14$, p=.049), having a high school diploma (${\beta}=.31$ p=.001), making less than 2-3 million won (${\beta}=-.19$, p=.036) per month, experiencing a sense of loss (${\beta}=-.22$, p=.003) and family support (${\beta}=.44$, p<.001). The total explanatory power amounted to 53.8% (F=14.83, p<.001, $AdjR^2=.54$). Conclusion: Educational intervention programs for breast cancer surgery patients that improve resilience by reducing the sense of loss and increasing family support must be developed.