• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.13 no.1
• /
• pp.93-102
• /
• 2010

Greening sometimes fails because its method is not suitable for various site conditions, therefore the trend of selecting a revegetation method in Korea today is through test construction. However, due to enlargement, complication and diversification of domestic construction businesses, the importance of VE is gradually increasing as effective efforts over a whole life-cycle to obtain goals such as quality improvement and cost reduction, and not only quality and economic efficiency but also substantiality need to be considered in comparing revegetation methods. For this study, Sungnam~Janghowon (area1), where comparatively various slope revegetation methods are used, was selected the investigation site. The site was divided into three areas:blasting rock, ripping rock and earth sand. The revegetation methods used were six in the blasting rock area, five in the ripping rock area, and two in the earth sand region. 2007 monitoring data was analyzed, and Value (V) was calculated with LCC related ratio, and compared and contrasted with the evaluation of prior revegetation methods. Therefore it is believed that this analysis enables selection of the most appropriate method, unbiased towards one particular characteristic such as quality, vegetation growth and economy. When aiming for a durable effect, it shall be more efficient to select the most appropriate method focusing on LCC analysis, which deals with the economic aspect, as well as the design function aspect.

• Journal of Korean Institute of Industrial Engineers
• /
• v.41 no.1
• /
• pp.25-33
• /
• 2015

From the viewpoint of applications to manufacturing, data mining is a useful method to find the meaningful knowledge or information about states of processes. But the data from manufacturing processes usually have two characteristics which are multicollinearity and imbalance distribution of data. Two characteristics are main causes which make bias to classification rules and select wrong variables as important variables. In the paper, we propose a new data mining procedure to solve the problem. First, to determine candidate variables, we propose the multiple hypothesis test. Second, to make unbiased classification rules, we propose the decision tree learning method with different weights for each category of quality variable. The experimental result with a real PDP (Plasma display panel) manufacturing data shows that the proposed procedure can make better information than other data mining procedures.

• Communications for Statistical Applications and Methods
• /
• v.24 no.4
• /
• pp.383-396
• /
• 2017

The model in our approach assumes that computer responses are a realization of a Gaussian processes superimposed on a regression model called a Gaussian process regression model (GPRM). Selecting a subset of variables or building a good reduced model in classical regression is an important process to identify variables influential to responses and for further analysis such as prediction or classification. One reason to select some variables in the prediction aspect is to prevent the over-fitting or under-fitting to data. The same reasoning and approach can be applicable to GPRM. However, only a few works on the variable selection in GPRM were done. In this paper, we propose a new algorithm to build a good prediction model among some GPRMs. It is a post-work of the algorithm that includes the Welch method suggested by previous researchers. The proposed algorithms select some non-zero regression coefficients (${\beta}^{\prime}s$) using forward and backward methods along with the Lasso guided approach. During this process, the fixed were covariance parameters (${\theta}^{\prime}s$) that were pre-selected by the Welch algorithm. We illustrated the superiority of our proposed models over the Welch method and non-selection models using four test functions and one real data example. Future extensions are also discussed.

• ETRI Journal
• /
• v.42 no.6
• /
• pp.886-898
• /
• 2020

Unusual data patterns or outliers can be generated because of human errors, incorrect measurements, or malicious activities. Detecting outliers is a difficult task that requires complex ensembles. An ideal outlier detection ensemble should consider the strengths of individual base detectors while carefully combining their outputs to create a strong overall ensemble and achieve unbiased accuracy with minimal variance. Selecting and combining the outputs of dissimilar base learners is a challenging task. This paper proposes a model that utilizes heterogeneous base learners. It adaptively boosts the outcomes of preceding learners in the first phase by assigning weights and identifying high-performing learners based on their local domains, and then carefully fuses their outcomes in the second phase to improve overall accuracy. Experimental results from 10 benchmark datasets are used to train and test the proposed model. To investigate its accuracy in terms of separating outliers from inliers, the proposed model is tested and evaluated using accuracy metrics. The analyzed data are presented as crosstabs and percentages, followed by a descriptive method for synthesis and interpretation.

• Structural Engineering and Mechanics
• /
• v.77 no.3
• /
• pp.417-439
• /
• 2021

The geometric nonlinearity has been successfully integrated with the design of steel structural system. Thus, the tubular lattice girder, one application of steel structural systems have already been optimized to obtain an economic design following the completion of computationally expensive design procedure. In order to decrease its computing cost, this study proposes to employ five multi-objective metaheuristics for the design optimization of geometrically nonlinear tubular lattice girder. Then, the employed multi-objective optimization algorithms (MOAs), NSGAII, PESAII, SPEAII, AbYSS and MoCell are evaluated considering their computing performances. For an unbiased evaluation of their computing performance, a tubular lattice girder with varying size-shape-topology and a benchmark truss design with 17 members are not only optimized considering the geometrically nonlinear behavior, but three benchmark mathematical functions along with the four benchmark linear design problems are also included for the comparison purpose. The proposed experimental study is carried out by use of an intelligent optimization tool named JMetal v5.10. According to the quantitative results of employed quality indicators with respect to a statistical analysis test, MoCell is resulted with an achievement of showing better computing performance compared to other four MOAs. Consequently, MoCell is suggested as an optimization tool for the design of geometrically nonlinear tubular lattice girder than the other employed MOAs.

• Journal of the Korean Society of School Health
• /
• v.31 no.2
• /
• pp.79-87
• /
• 2018

Purpose: The purpose of this study was to investigate the relationship between the risk of school bullying victimization and the risk of Internet gaming disorder according to gender in adolescents. Methods: The data of 1,920 middle school students collected at the baseline of the Internet user Cohort for Unbiased Recognition of gaming disorder in Early Adolescence (iCURE) study were analyzed. For statistical analysis, $x^2$ test, t-test and stratified multiple logistic regression analysis were conducted using SAS 9.4. Results: The prevalence rate of Internet gaming disorder of middle school boys was greater than that of girls (Boys: 9.9%, Girls: 6.2%). The greater the risk of school bullying victimization, the greater both the risk of Internet gaming disorder and the average daily time spent on Internet gaming. In girls, the relationship between the experience of being bullied in school and Internet gaming disorder was not statistically significant. However, the boys who had been bullied in school were 3.2 times more vulnerable to the risk of Internet gaming disorder than those without such experience (95% CI: 1.135-8.779). Conclusions: When considering interventions for Internet gaming disorder for adolescents, bullying victimization should be taken into account as well. Particularly, relieving stress related to bullying victimization can be important for boys with Internet gaming disorder.

• Journal of Korean Society on Water Environment
• /
• v.25 no.5
• /
• pp.792-802
• /
• 2009

The objective of this study is to construct the watershed management system with link of the non-point sources model and to estimate delivery ratio duration curves for various pollutants. For the total water pollution load management system, non-point source model should be performed with the study of the characteristic about non-point sources and loadings of non-point source and the allotment of pollutant in each area. In this study, daily flow rates and delivered pollutant loads of Nakdong river basin are simulated with modified TANK model and minimum variance unbiased estimator and SWAT model. Based on the simulation results, flow duration curves, load duration curves, and delivery ratio duration curves have been established. Then GIS analysis is performed to obtain several hydrological geomorphic characteristics such as watershed area, stream length, watershed slope and runoff curve number. As a result, the SWAT simulation results show good agreements in terms of discharge, BOD, TN, TP but for more exact simulation should be kept studying about variables and parameters which are needed for simulation. And as a result of the characteristic discharges, pollutants loading with the runoff and delivery ratios, non-point sources effects were higher than point sources effects in the small-scale test bed of Nakdong river basin.

• Animal Bioscience
• /
• v.36 no.9
• /
• pp.1327-1335
• /
• 2023

Objective: When evaluating individuals with the same parent and no phenotype by pedigree best linear unbiased prediction (BLUP), it is difficult to explain carcass grade difference and select individuals because they have the same value in pedigree BLUP (PBLUP). However, single step GBLUP (ssGBLUP), which can estimate the breeding value suitable for the individual by adding genotype, is more accurate than the existing method. Methods: The breeding value and accuracy were estimated with pedigree BLUP and ssGBLUP using pedigree and genotype of 408 Hanwoo cattle from 16 families with the same parent among siblings produced by fertilized egg transplantation. A total of 14,225 Hanwoo cattle with pedigree, genotype and phenotype were used as the reference population. PBLUP obtained estimated breeding value (EBV) using the pedigree of the test and reference populations, and ssGBLUP obtained genomic EBV (GEBV) after constructing and H-matrix by integrating the pedigree and genotype of the test and reference populations. Results: For all traits, the accuracy of GEBV using ssGBLUP is 0.18 to 0.20 higher than the accuracy of EBV obtained with PBLUP. Comparison of EBV and GEBV of individuals without phenotype, since the value of EBV is estimated based on expected values of alleles passed down from common ancestors. It does not take Mendelian sampling into consideration, so the EBV of all individuals within the same family is estimated to be the same value. However, GEBV makes estimating true kinship coefficient based on different genotypes of individuals possible, so GEBV that corresponds to each individual is estimated rather than a uniform GEBV for each individual. Conclusion: Since Hanwoo cows bred through embryo transfer have a high possibility of having the same parent, if ssGBLUP after adding genotype is used, estimating true kinship coefficient corresponding to each individual becomes possible, allowing for more accurate estimation of breeding value.

• Asian-Australasian Journal of Animal Sciences
• /
• v.33 no.3
• /
• pp.382-389
• /
• 2020

Objective: This study was conducted to test the efficiency of genomic selection for milk production traits in a Korean Holstein cattle population. Methods: A total of 506,481 milk production records from 293,855 animals (2,090 heads with single nucleotide polymorphism information) were used to estimate breeding value by single step best linear unbiased prediction. Results: The heritability estimates for milk, fat, and protein yields in the first parity were 0.28, 0.26, and 0.23, respectively. As the parity increased, the heritability decreased for all milk production traits. The estimated generation intervals of sire for the production of bulls (L_SB) and that for the production of cows (L_SC) were 7.9 and 8.1 years, respectively, and the estimated generation intervals of dams for the production of bulls (L_DB) and cows (L_DC) were 4.9 and 4.2 years, respectively. In the overall data set, the reliability of genomic estimated breeding value (GEBV) increased by 9% on average over that of estimated breeding value (EBV), and increased by 7% in cows with test records, about 4% in bulls with progeny records, and 13% in heifers without test records. The difference in the reliability between GEBV and EBV was especially significant for the data from young bulls, i.e. 17% on average for milk (39% vs 22%), fat (39% vs 22%), and protein (37% vs 22%) yields, respectively. When selected for the milk yield using GEBV, the genetic gain increased about 7.1% over the gain with the EBV in the cows with test records, and by 2.9% in bulls with progeny records, while the genetic gain increased by about 24.2% in heifers without test records and by 35% in young bulls without progeny records. Conclusion: More genetic gains can be expected through the use of GEBV than EBV, and genomic selection was more effective in the selection of young bulls and heifers without test records.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.158-163
• /
• 2005

Competitive electricity markets necessitate equitable methods for allocating transmission usage in order to set transmission usage charges and congestion charges in an unbiased and an open-accessed basis. So in competitive markets it is usually necessary to trace the contribution of each participant to line usage, congestion charges and transmission losses, and then to calculate charges based on these contributions. A UPFC offers flexible power system control, and has the powerful advantage of providing, simultaneously and independently, real-time control of voltage, impedance and phase angle, which are the basic power system parameters on which sys-tem performance depends. Therefore, UPFC can be used efficiently and flexibly to optimize line utilization and increase system capability and to enhance transmission stability and dampen system oscillations. In this paper, a mathematical approach to allocate the contributions of system users and UPFCs to transmission system usage is presented. The paper uses a dc-based load flow modeling of UPFC-inserted transmission lines in which the injection model of the UPFC is used. The relationships presented in the paper showed modified distribution factors that modeled impact of utilizing UPFCs on line flows and system usage. The derived relationships show how bus voltage angles are attributed to each of changes in generation, injections of UPFC, and changes in admittance matrix caused by inserting UPFCs in lines. The relationships derived are applied to two test systems. The results illustrate how transmission usage would be affected when UPFC is utilized. The relationships derived can be adopted for the purpose of allocating usage and payments to users of transmission network and owners of UPFCs used in the network. The relationships can be modified or extended for other control devices.