• Communications for Statistical Applications and Methods
• /
• v.22 no.6
• /
• pp.625-637
• /
• 2015

We develop a Bayesian clustering procedure based on a Dirichlet process prior with cluster specific random effects. Gibbs sampling of a normal mixture of linear mixed regressions with a Dirichlet process was implemented to calculate posterior probabilities when the number of clusters was unknown. Our approach (unlike its counterparts) provides simultaneous partitioning and parameter estimation with the computation of the classification probabilities. A Monte Carlo study of curve estimation results showed that the model was useful for function estimation. We find that the proposed Dirichlet process mixture model with cluster specific random effects detects clusters sensitively by combining vague edges into different clusters. Examples are given to show how these models perform on real data.

• Publications of The Korean Astronomical Society
• /
• v.30 no.2
• /
• pp.489-490
• /
• 2015

We present population synthesis models for the calcium II triplet (CaT), currently the most popular metallicity indicator, based on high-resolution empirical spectral energy distributions (SEDs). Our new CaT models, based on empirical SEDs, show a linear correlation below [Fe/H] ~ -0.5, but the linear relation breaks down in the metal-rich regime by converging to the same equivalent width. This relation shows good agreement with the observed CaT of globular clusters (GCs) in NGC 1407 and the Milky Way. However, a model based on theoretical SEDs does not show this feature of the CaT and fails to reproduce observed GCs in the metal-rich regime. This linear relation may cause inaccurate metallicity determination for metal-rich stellar populations. We have also confirmed that the effect of horizontal-branch stars on the CaT is almost negligible in models based on both empirical and theoretical SEDs. Our new empirical model may explain the difference between the color distributions and CaT distributions of GCs in various early-type galaxies. Based on our model, we claim that the CaT is not a good metallicity indicator for simple stellar populations in the metal-rich regime.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.37 no.2
• /
• pp.55-64
• /
• 2000

In this paper, we propose an efficient datapath placement algorithm to minimize track density. Here, we consider each datapath element as a cluster, and merge the most strongly connected two clusters to a new cluster until only one cluster remains. As nodes in the two clusters to be merged are already linearly ordered respectively, we can merge two clusters with connecting them. The proposed algorithm produces circular linear ordering by connecting starting point and end point of the final cluster, and n different linear ordering by cutting between two contiguous elements of the circular linear ordering. Among the n different linear ordering, the linear ordering to minimize track density is final solution. In this paper, we show and utilize that if two clusters are strongly connected in a graph, the inner product of the corresponding vectors mapped in d-dimensional space using spectral method is maximum. Compared with previous datapath placement algorithm GA/S $A^{［2］}$, the proposed algorithm gives similar results with much less computation time.

• Korean Journal of Materials Research
• /
• v.1 no.3
• /
• pp.139-144
• /
• 1991

The geometries and energies of $C_2-C_5$ clusters have been calculated using simple semiempirical potential energy functions. The results of the calculations show that the most stable structure of the $C_2-C_5$ clusters is linear and that not only the rhombic $C_4$ but also the Y-shaped $C_4$ hale similar energy to the linear $C_4$.

• Journal of Astronomy and Space Sciences
• /
• v.14 no.2
• /
• pp.179-193
• /
• 1997

We use B band CCD images to investigate the surface brightness distributions and dynamical properties of 29 Galactic globular clusters. Model fits suggest that 22 clusters show King type surface brightness profiles, while 7 clusters are characterized by power law cusp profiles. For the King type clusters, concentration parameters $(c=log(r_{t}/r_{c}))$ range from 1.20 to 2.10, and core radii are 0.4 to 1.9 pc. The mean value of power law slopes of 7 cuspy clusters was estimated as ${alpha}=1.011{\pm}0.065$. Total masses of King type globular clusters are in the range of $1.7{ imes}10^4M_{odot}$ to $1.0{\times}106M_{\odot}$ with a mean of $1.7{\times}10^5M_{\odot}$. A significant positive correlation between mass and mass-to-light ratio of King type globular clusters has been confirmed with a Pearson's correlation coefficient r = 0.52 and a confidence level of 99%. Our data also confirm a linear relation between total mass and absolute magnitude of King type globular clusters.

• Communications for Statistical Applications and Methods
• /
• v.14 no.1
• /
• pp.205-213
• /
• 2007

In this paper, the normal mixture model subjected to general linear restriction for component-means based on linear regression is proposed, and its fitting method by EM algorithm and Lagrange multiplier is provided. This model is applied to gene clustering of microarray expression data, which demonstrates it has very good performances for real data set. This model also allows to obtain the clusters that an analyst wants to find out in the fashion that the hypothesis for component-means is represented by the design matrices and the linear restriction matrices.

• Journal of Korean Critical Care Nursing
• /
• v.12 no.1
• /
• pp.1-12
• /
• 2019

Purpose : It is widely accepted that addressing multiple symptoms together is the preferred approach in assessment and intervention and results in reduced negative patient outcomes. Yet, there are few studies examining symptom clusters and their impacts on quality of life longitudinally in patients after heart valve surgery. Methods: A total of 101 patients were recruited from a tertiary hospital and were administered questionnaires (at 3, 6, and 10 weeks after the surgery) assessing participants' characteristics, cardiac symptoms, and quality of life. Factor analysis was used to identify symptom clusters. Hierarchical multiple linear regression was used to predict quality of life. Results: Participants were predominantly 70-years old or more with a mean age of 64.34. The two symptom clusters at 3 weeks after the surgery with education, gender, and occupation accounted for 76.3% of variance in quality of life. Conclusion: Symptom clusters containing various physical and psychological symptoms in patients after the surgery affected quality of life, and the relationship was significant at 3 weeks after the surgery. Because symptom clusters were identified in all three recovery periods, nurses need to acknowledge these clusters, rather than each symptom separately, and to utilize these in providing care and education and in promoting quality of life in these patients.

• Korean Journal of Materials Research
• /
• v.2 no.2
• /
• pp.95-100
• /
• 1992

The bond lengths, bond angles and energies of $C_6$ clusters are calculated by using the simulation technique based on some semiempirical potential energy functions, with the result that the equilibrium configuration of $C_6$ is a regular equilateral hexagon. The same simulations have been performed for linear and cyclic forms of $C_7-C_10$ clusters and the results show that the regular heptagonal, octagonal, nonagonal and decagonal forms are more stable than the linear forms of the respective clusters. Comparison of this result with the present author's previous result that the most stable configurations of $C_2-C_5$ are linear implies that the most stable configuration of the cluster changes from one-dimension al to two-dimensional one as the cluster size increases and that the critical cluster size is $C_6$.

• Korean Journal of Human Ecology
• /
• v.8 no.2
• /
• pp.313-325
• /
• 1999

This study aimed to classify the data of children's bodies into several clusters by principal component scores that were extracted through the factor analysis in the former study, and to describe the distribution and body characteristics of the clusters. The sample was 308 elementary school children aged from 6 to 8 and the anthropometric measurements were performed indirectly from the photographs of the subjects, which was the same as the first analysis. The data were analysed statistically using SPSSWIN Ver. 8.0. Through the statistical analysis, 3 clusters were obtained from the data. The first cluster distributed more in the children aged 7 and 8 than in the children aged 6. The somatotype of this group was the tallest among the three groups, and they were the most developed group compared to the two other groups in lateral component as well as in linear component. The second cluster group wasn't well developed in lateral components, and had lowest level in Rohrer Index, so this group had thin figures compared to the other groups. The third cluster revealed dominant distribution in the group aged 6, and the group had the least developed linear components but higher level in Rohrer Index. Each cluster group revealed peculiar somatotype that was dominant in one group but rarely in other cluster groups. Lateral views of these characteristics were showed using the average of the measurements of clusters.

• Journal of Communications and Networks
• /
• v.14 no.4
• /
• pp.443-450
• /
• 2012

As for cluster-based wireless sensor networks (WSNs), cluster lifetime is one of the most important subjects in recent researches. Besides reducing the energy consumptions of the clusters, it is necessary to make the clusters achieve equal lifetimes so that the whole network can survive longer. In this paper, we focus on the cluster lifetimes in multi-hop WSNs with cooperative multi-input single-output scheme. With a simplified model of multi-hop WSNs, we change the transmission schemes, the sizes and transmission distances of clusters to investigate their effects on the cluster lifetimes. Furthermore, linear and uniform data aggregations are considered in our model. As a result, we analyze the cluster lifetimes in different situations and discuss the requirements on the sizes and transmission distances of clusters for equal lifetimes.