• Journal of Preventive Medicine and Public Health
• /
• 제33권1호
• /
• pp.51-55
• /
• 2000

Objectives . In large-scale field trials, randomization by cluster is frequently used because of the administrative convenience, a desire to reduce the effect of treatment contamination, and the need to avoid ethical issues that might of otherwise arise. Cluster randomization trials are experiments in which intact social unit, e.g., families, schools, cities, rather than independent individuals are randomly allocated to intervention groups. The positive correlation among responses of subjects from the same cluster is in matter in cluster randomization. This thesis is to compare the results of three randomization methods by standard error of estimator of treatment effect. Methods : We simulated cholesterol data varing the size of the cluster and the level of the correlation in clusters and analyzed the effect of cholesterol-lowering agent. Results : In intra-cluster randomization the standard error of the estimator of treatment effect is smallest relative to that in inter-cluster randomization and that in individual randomization. Conclusions : Infra-cluster randomization is the most efficient in its standard error of estimator of treatment effect but other factor should be considered when selecting a specific randomization method.

• 한국전산유체공학회:학술대회논문집
• /
• 한국전산유체공학회 2001년도 추계 학술대회논문집
• /
• pp.58-62
• /
• 2001

Aparallel computing environment was constructed based on Windows 2000 operating system. This cluster was configured using Fast-Ethernet system to hold up together the clients within a network domain. For the parallel computation, MPI implements for Windows such as MPICH.NT.1.2.2 and MP-MPICHNT.1.2 were used with Compaq Visual Fortran compiler which produce a well optimized executives for x86 systems. The evaluation of this cluster performance was carried out using a preconditioned Navier-Stokes code for the 2D analysis of a compressible and viscous flow around a compressor blade. The parallel performance was examined in comparison with those of Linux clusters studied previously by changing a number of processors, problem size and MPI libraries. The result from the test problems presents that parallel performance of the low cost Fast-Ethernet Windows cluster is superior to that of a Linux cluster of similar configuration and is comparable to that of a Myrinet cluster.

• Communications for Statistical Applications and Methods
• /
• 제23권1호
• /
• pp.47-55
• /
• 2016

Stratified cluster sampling has been widely used for effective parameter estimations due to reductions in time and cost. The probability proportional to size (PPS) sampling method is used when the number of cluster element are significantly different. However, simple random sampling (SRS) is commonly used for simplicity if the number of cluster elements are almost the same. Also it is known that the ratio estimator produces a good performance when the total number of population elements is known. However, the two stage cluster estimator should be used if the total number of elements in population is neither known nor accurate. In this study we suggest a composite estimator by combining the ratio estimator and the two stage cluster estimator to obtain a better estimate under a certain population circumstance. Simulation studies are conducted to compare the superiority of the suggested estimator with two other estimators.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 ITC-CSCC -1
• /
• pp.314-317
• /
• 2000

Cluster system provides attractive scalability in terms of computation power and memory size. With the advances in high speed computer network technology, cluster systems are becoming increasingly competitive compared to expensive parallel machines. In parallel processing program, each task load is difficult to predict before running the program and each task is interdependent each other in many ways. Load imbalancing induces an obstacle to system performance. Most of researches in load balancing were concerned with distributed system but researches in cluster system are few. In cluster system, the dynamic load balancing algorithm which evaluates each processor's load in runtime is purpose that the load of each node are evenly distributed. But, if communication cost or node complexity becomes high, it is not effective method for all nodes to attend load balancing process. In that circumstances, it is good to reduce the number of node which attend to load balancing process. We have modeled cluster systems and proposed marginal dynamic load balancing algorithms suitable for that circumstances.

• 한국컴퓨터정보학회논문지
• /
• 제24권1호
• /
• pp.1-7
• /
• 2019

In this paper, we implemented a Wi-Fi based cluster system using raspberry pi for multidisciplinary education. The cluster implementation on the desktop was more difficult to maintain the complexity, big size, high price, power consumption as the number of nodes increased. In this paper, we implemented a cluster using Raspberry Pi, which is developed for educational purposes, to reduce the cost of connecting nodes. In addition, the complexity of system construction is reduced by replacing the connection between each node with Wi-Fi. Also, the inconvenience of configuration due to node increase was reduced. It is expected that the implementation of the cluster will be a good alternative in the educational environment where distributed processing and parallel processing are performed in the embedded environment. Also, it is confirmed that it can be applied to the multidisciplinary education.

• 천문학회보
• /
• 제40권1호
• /
• pp.76.1-76.1
• /
• 2015

We study the environmental dependence of the mass-size relation especially for the most massive early type galaxies (M>$10^{11.2}M_{\odot}$) in the redshift range 0.15~0.25. As a measure of the environment, galaxy number densities are measured by the $10^{th}$ nearest galaxies within 7000km/s from galaxies with spectroscopic redshifts. We find that galaxies more massive than $10^{11.6}M_{\odot}$ show the environmental dependence in the mass-size relation. The galaxies with M>$10^{11.6}M_{\odot}$ located in the densest, cluster like environment have larger sizes than their counterparts located in a low dense environment. We also find that this environmental dependence of the mass-size relation originates from the brightest cluster galaxies (BCG) rather than non-BCG galaxies. Our result can be explained with a hierarchical growth of the most massive galaxies through dissipation-less merger in dense environments.

• 천문학회보
• /
• 제40권2호
• /
• pp.29.1-29.1
• /
• 2015

We study the environmental dependence of the mass-size relation for the most massive early type galaxies (M > $10^{10.7}M_{\odot}$) in the redshift range 0.10~0.15. As a measure of the environment, galaxy number densities are measured by the $10^{th}$ nearest galaxies within 6500km/s from galaxies with spectroscopic redshifts. The sizes of galaxies are measured by non-parametric method. We find that galaxies more massive than $10^{11.1}M_{\odot}$ show the environmental dependence in the mass-size relation. The galaxies with M > $10^{11.1}M_{\odot}$ located in the densest, cluster like environment have larger sizes and extended surface brightness profiles than their counterparts located in a low dense environment. We also find that the environmental dependence of the mass-size relation is more significant for the brightest cluster galaxies (BCGs) than non-BCGs. Our result can be explained with a hierarchical growth of the most massive galaxies through dissipation-less merger in dense environment.

• 천문학회보
• /
• 제41권1호
• /
• pp.69.2-69.2
• /
• 2016

We study the environmental dependence of the mass-size relation for the most massive early type galaxies (M>$10^{10.7}M_{\odot}$) in the redshift range 0.10~0.15. The sizes of galaxies are measured by non-parametric method. We find that galaxies more massive than $10^{11.1}M_{\odot}$ show the environmental dependence in the mass-size relation. The galaxies with M>$10^{11.1}M_{\odot}$ located in the densest, cluster like environment have larger sizes and extended surface brightness profiles than their counterparts located in a low dense environment. We also find that the environmental dependence of the mass-size relation is more significant for the brightest cluster galaxies (BCGs) than non-BCGs. We use the semi analytic galaxy formation simulation based on the Millennium 1 Simulation for interpretation. Our result can be explained with a hierarchical growth of the most massive galaxies through dissipation-less merger in dense environment.

• 한국의류학회지
• /
• 제22권2호
• /
• pp.203-214
• /
• 1998

The purpose of this study is to classify the upper body of women into several kinds of somatotypes, using the method of Surgical Tape and making their shells. The subjects are 53 females 30 to 39 years-old. Fifty-three anthropometric data are measured per shell of bodysurface; six somatotype factors are obtained through principal component analysis and orthogonal rotation by the method of Varimax, Somatotype of women's upper body is achieved by cluster analysis, using the standardized factor score as an independent variable and the FASTCLUS of SAS by Kmeans. The results are as follows: 1. The number of the factors which explain the somatotype is six and those factors comprise 76.12 percent of total variance. Factor 1: related to the size of shape in the front of upper body Factor S: related to the size of shape in the back of upper body Factor 3: related to the type of the upper chest over the chest circumference line Factor 4: related to the length of·the upper body Factor 5: related to the part of the neck Factor 6: related to the type of the lower chest under the chest circumference line 2. Cluster analysis results in classification of upper body into five clusters. Cluster L: the length is the largest and the circumference is small. The part of waist is the largest and widest among surface areas. Cluster 2: Slender body line from chest to waist is characteristic. The length is longer. The part of upper and lower chest is larger among surface areas. Cluster S: the circumference is the smallest and armhole is small. The length and surface area are small. Cluster 4: the circumference and armhole is the largest. The length is the smallest. Cluster 5: the circumference is average and the length is a little long. The body line(silhouette) from chest to waist is curved slightly.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권9호
• /
• pp.4228-4247
• /
• 2018

Many methods have been developed for the vehicles to create clusters in vehicular ad hoc networks (VANETs). Usually, nodes are vehicles in the VANETs, and they are dynamic in nature. Clusters of vehicles are made for making the communication between the network nodes. Cluster Heads (CHs) are selected in each cluster for managing the whole cluster. This CH maintains the communication in the same cluster and with outside the other cluster. The lifetime of the cluster should be longer for increasing the performance of the network. Meanwhile, lesser the CH's in the network also lead to efficient communication in the VANETs. In this paper, a novel algorithm for clustering which is based on the social behavior of Gray Wolf Optimization (GWO) for VANET named as Intelligent Clustering using Gray Wolf Optimization (ICGWO) is proposed. This clustering based algorithm provides the optimized solution for smooth and robust communication in the VANETs. The key parameters of proposed algorithm are grid size, load balance factor (LBF), the speed of the nodes, directions and transmission range. The ICGWO is compared with the well-known meta-heuristics, Multi-Objective Particle Swarm Optimization (MOPSO) and Comprehensive Learning Particle Swarm Optimization (CLPSO) for clustering in VANETs. Experiments are performed by varying the key parameters of the ICGWO, for measuring the effectiveness of the proposed algorithm. These parameters include grid sizes, transmission ranges, and a number of nodes. The effectiveness of the proposed algorithm is evaluated in terms of optimization of number of cluster with respect to transmission range, grid size and number of nodes. ICGWO selects the 10% of the nodes as CHs where as CLPSO and MOPSO selects the 13% and 14% respectively.