• Journal of Life Science
• /
• v.19 no.10
• /
• pp.1403-1409
• /
• 2009

The purpose of this study was to compare Obesity Age (OA) and chronological age, to calculate Obesity Age (OA), which gauges the state of obesity, and to analyze presented factors of obesity using expectant factors on middle-aged obese women. The subjects were one hundred twenty seven middle-aged obese women ($49.6\pm7.3$ yr, BMI $29.41\pm2.9$, fat $36.8\pm4.6%$) who participated in different weight loss programs three times. The body composition, physical fitness, blood pressure and blood were measured before the weight loss programs. Informed consent was obtained from all subjects before enrollment in the study. The regression equation is as follows: (1) OAS (Obesity Age Score)=$0.106*X_1+0.035*X_2+0.048*X_3+0.041*X_4+0.003*X_5-0.037*X_6-10.667$ ($X_1$: BMI, $X_2$: weight, $X_3$: %fat, $X_4$: WC, $X_5$: TG, $X_6$: $VO_{2max}$), (2) OA (Obesity Age)=7.3*OAS+49.6*(-1), (3) Z (correction factor)=(CA-49.6)(1-0.03), (4) OAc (corrected Obesity Age)=1.03*CA-7.3*OAS+1.47. The comparison of corrected Obesity Age (OAc) and chronological age did not have any differences, and the average of the OAc was close to chronological age. The correlation coefficient between the OAc and chronological age was r=0.724 (p<0.05). The equation can be utilized for middle-aged obese women, because it could evaluate the obesity-related factors by including BMI, body weight, %fat, waist circumference, triglycerides and $VO_{2max}$.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.45 no.3
• /
• pp.342-351
• /
• 2016

This study was carried out to develop an optimized analytical method using high-performance liquid chromatography (HPLC) applied to canthaxanthin, which is not yet designated as a food colorant in Korea, as well as to perform validation and uncertainty evaluation of this method. Official methods of AOAC, UK, and Japan with HPLC-UV detection were evaluated for the analysis of canthaxanthin by comparison of linearity, resolution, selectivity, limit of detection (LOD), limit of quantitation (LOQ), accuracy, precision, recovery, inter-laboratory tests, and uncertainty measurement. The calibration curves showed high linearity with an $R_2$ value of over 0.999 for canthaxanthin standard solutions in all three official methods. The official method of Japan exhibited the best results in terms of resolution and selectivity, including the lowest LOD and LOQ. The average coefficients of variation were calculated as less than five of three institutes with a precision value less than 1, accuracy near 100%, and recovery ratio between $100{\pm}10%$. The expanded uncertainty for canthaxanthin was estimated to be $39.5{\pm}5.29mg/kg$ (95% confidence level, k=2), and the uncertainty of measurement was 13.4%. In this study, official methods of canthaxanthin were compared and the validities verified. The results will be further applied to establish an authorized analytical method for canthaxanthin in Korea.

• Journal of KIISE
• /
• v.44 no.9
• /
• pp.954-965
• /
• 2017

A stable power supply is very important for the maintenance and operation of the power infrastructure. Accurate power consumption prediction is therefore needed. In particular, a university campus is an institution with one of the highest power consumptions and tends to have a wide variation of electrical load depending on time and environment. For this reason, a model that can accurately predict power consumption is required for the effective operation of the power system. The disadvantage of the existing time series prediction technique is that the prediction performance is greatly degraded because the width of the prediction interval increases as the difference between the learning time and the prediction time increases. In this paper, we first classify power data with similar time series patterns considering the date, day of the week, holiday, and semester. Next, each ARIMA model is constructed based on the classified data set and a daily power consumption forecasting method of the university campus is proposed through the time series cross-validation of the predicted time. In order to evaluate the accuracy of the prediction, we confirmed the validity of the proposed method by applying performance indicators.

• Journal of the Korean Society of Groundwater Environment
• /
• v.2 no.2
• /
• pp.58-63
• /
• 1995

Cokriging was applied for the estimation of the water levels of the basal leachate and the surrounding groundwater at the Nanjido waste landfill area. When the groundwater level is estimated at the high relief area, it makes a good result to use the data of groundwater level and elevation simultaneously because groundwater level is correlated with topography. This study determined the best semivariogram model of 87 groundwater levels and 144 elevations through cross validation test, and produced the contour maps of groundwater levels using ordinary kriging and universal kiging. Two contour maps don't make big difference at the waste site because this area has a large number of groundwater level data. However, they show big difference at the upper left part of the study area because this area has high relief and a small number of sample data. Their difference is also found at the south area near the Han river. When the topography is considered for the both areas, the contour map of cokriging is thought to be closer to the real groundwater distribution than that of kriging.

• The Korean Journal of Applied Statistics
• /
• v.26 no.3
• /
• pp.441-452
• /
• 2013

This study analyzed the relationship between key film and a box office record success factors based on movies released in the first quarter of 2013 in Korea. An over-fitting problem can happen if there are too many explanatory variables inserted to regression model; in addition, there is a risk that the estimator is instable when there is multi-collinearity among the explanatory variables. For this reason, optimal variable selection based on high explanatory variables in box-office performance is of importance. Among the numerous ways to select variables, LASSO estimation applied by a generalized linear model has the smallest prediction error that can efficiently and quickly find variables with the highest explanatory power to box-office performance in order.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.4
• /
• pp.739-748
• /
• 2012

In this paper we propose a mixed effects least squares support vector machine (LS-SVM) for the censored data which are observed from different groups. We use weights by which the randomly right censoring is taken into account in the nonlinear regression. The weights are formed with Kaplan-Meier estimates of censoring distribution. In the proposed model a random effects term representing inter-group variation is included. Furthermore generalized cross validation function is proposed for the selection of the optimal values of hyper-parameters. Experimental results are then presented which indicate the performance of the proposed LS-SVM by comparing with a standard LS-SVM for the censored data.

• Journal of KIISE:Software and Applications
• /
• v.34 no.7
• /
• pp.595-602
• /
• 2007

This paper proposes a method that discriminates signal peptide and predicts the cleavage site of the secretory proteins cleaved by the signal peptidase I. The preprocessing stage uses hydrophobicity scales of amino acids in order to predict the presence of signal sequence and the cleavage site. The preprocessing enhances the performance of the prediction method by eliminating the non-secretory proteins in the early stage of prediction. for the effective use of support vector machine for the signal sequence prediction, the biologically relevant distance between the amino acid sequences is defined by using the hydrophobicity and substitution matrix; the hydrophobicity can be used to Predict the location of amino acid in a cell and the substitution matrix represents the evolutionary relationships of amino acids. The proposed method showed 98.9% discrimination rates from signal sequences and 88% correct rate of the cleavage site prediction on Swiss-Prot release 50 protein database using the 5-fold-cross-validation. In the comparison tests, the proposed method has performed significantly better than other prediction methods.

• Journal of Industrial Convergence
• /
• v.20 no.12
• /
• pp.179-186
• /
• 2022

This study aims to develop a predictive model to suggest a used sales price to sellers and buyers when trading used tablet PCs. For model development, we analyzed the real used tablet PC transaction data and additionally collected detailed product information. We developed several predictive models and selected the best predictive model among them. Specifically, we considered a multiple linear regression model using the used sales price as a dependent variable and other variables in the integrated data as independent variables, a multiple linear regression model including interactions, and the models from stepwise variable selection in each model. The model with the best predictive performance was finally selected through cross-validation. Through this study, we can predict the sales price of used tablet PCs and suggest appropriate used sales prices to sellers and buyers.

• Korean Journal of Social Welfare
• /
• v.54
• /
• pp.5-33
• /
• 2003

This study is about the re-validation evaluation of the family Relationship Scale (FRS), developed to measure the family relationship in the social work practice. This study aims at re-validating the FRS, developed and validated in by Yang in 2001 for more general utilization. The sample was married mates and females residing in Seoul. For Face Validity, the content analysis was performed, and the FRS was re-validated in the dimensions of Love & Caring, Acceptance, and Recognition, positive affection, empathy, and autonomy and flexibility for each area. Internal reliability was .93, and internal consistency among three dimensions was 93%. For Empirical Validity, the Construct validity, the Criterion validity, and the Discriminant validity were performed. Construct Validity was validated through factor analyses. Commonalities for the factor analysis was 54%, and the factor loading for each factor was over .45. The confirmative factor analysis also confirmed the fitness of the scale. For Predictive Validity of Criterion Validity, regression analysis showed that the family stress scores became lower as the scores of the family relationship became higher; the discriminant analysis revealed that the family stress turned low ill tile group of high scores of family relationship. The Correlation analysis for Concurrent Validity was performed and the results showed the positive and significant relationship with a couple communication level (r=54) and a parent-child communication level (r=64). Life satisfaction and mental health level also revealed significantly positive correlation to prove Convergent Validity. Physical health level revealed a weak relationship with family relationship providing the evidence of Discriminant Validity. Discriminance was also proved by the analysis of variance with demographics. Thus, Cross Validation was confirmed the validation of the FRS through the various analyses with the married population. This study result improved the validity generalization of the Scale and verify the generalized usage of this sociometric scale in the field of social work practice.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.2
• /
• pp.349-359
• /
• 2017

In this paper, we conducted survival analyses by fitting the Cox proportional hazards model to stage III proximal colon cancer data obtained from the Surveillance, Epidemiology, and End Results program of the National Cancer Institute. We investigated the effect of covariates on the hazard function for death from proximal colon cancer in stage III with surgery performed and estimated the survival probability for a patient with specific covariates. We showed that the proportional hazards assumption is satisfied for covariates that were used to analyses, using a test based on the Schoenfeld residuals and plots of the Schoenfeld residuals and $log[-log\{{\hat{S}}(t)\}]$. We evaluated the model calibration and discriminatory accuracy by calibration plot and time-dependent area under the ROC curve, which were calculated using 10-fold cross validation.