• 한국대기환경학회지
• /
• 제24권3호
• /
• pp.300-309
• /
• 2008

Variations in PM10 concentration between 2000 and 2005 from 84 urban air monitoring stations operated by the government were analyzed. The K-means cluster analysis was attempted using annual average and the 99th percentile of daily averages as parameters. The results obtained by excluding Asian dust episode days were compared with those obtained by using all available data. In any cases, the cluster with the highest mean concentration was mostly composed of stations in Seoul and Gyeonggi. Annual average of the cluster with the highest mean concentration showed a distinct decreasing trend, but that excluding Asian dust episode days did not show such a trend. Without Asian dust episode days high concentrations of monthly averages in March and April were also not observed. The effect of Asian dust was more pronounced in the 99th percentile of daily averages. The 99th percentile of daily averages of the cluster with the highest mean concentration was the highest in June following downs in April and May.

• 한국시뮬레이션학회:학술대회논문집
• /
• 한국시뮬레이션학회 2001년도 The Seoul International Simulation Conference
• /
• pp.99-102
• /
• 2001

We propose a PC cluster using general-purpose microprocessors and a high-speed network for simulating ANN (Artificial Neural Network) processes on Linux OS. We apply this cluster to intelligent information processing such as ANN simulation. The elapsed time for simulating ANNs can be reduced from 7,295 seconds by a PE (Processing Element) to 1,226 seconds by six PEs. The reliability of a pattern-classification using ANNs can be improved by the proposed ANN, Randomized-ANN. In order to generate a Randomized-ANN, we choose three ANNs and combine the output results from three huts by means of logical AND. Results are as follows: The mean correct answer rate is 94.4%, the mean wrong answer rate is only 0.1 %, and the mean unknown answer rate is 5.5 %. We make sure that Randomized-ANN approach reduces the mean wrong answer rate within a tenth part and improves the reliability of Japanese coin classification.

• 동아시아식생활학회지
• /
• 제26권5호
• /
• pp.400-406
• /
• 2016

This study used K-mean cluster analysis to evaluate the preference and satisfaction according to consumption behavior of naturally fermented bread products among customers residing in the Seoul area. Naturally fermented bread products were best recognized as "great nutrients for good health" ($3.91{\pm}0.87$). The preference for naturally fermented bread products was due to "good taste and flavor" ($3.39{\pm}0.95$), and customers with "intention to purchase" showed a mean of $3.21{\pm}0.94$. The overall satisfaction for naturally fermented bread products was $3.26{\pm}0.75$. Among the specific categories that contributed to this overall satisfaction, "quality" showed the highest satisfaction with $3.43{\pm}0.77$, whereas "price" ($2.77{\pm}0.76$) and "variety" ($2.77{\pm}0.75$) exhibited the lowest. Among the items to modify for naturally fermented bread products, "variety" was the most important item (21.8%), followed by "lower price" and "convenience of purchase" at 19.7% and 17.9%, respectively. In K-mean cluster analysis, customers who frequently visited the bakery and purchased naturally fermented bread products (cluster 1) expressed strong preference, satisfaction, and consumption behavior. Furthermore, these customers expressed high satisfaction in "quality", "convenience of purchase", and "variety" of naturally fermented bread products.

• 한국의류학회지
• /
• 제10권3호
• /
• pp.37-48
• /
• 1986

The purpose of this study was to analyze 'the natural groupings' of subjects in order to classify highly similar somatotype for clothing construction. The sample for the study was drawn randomly out of senior high school boys in Seoul urban area. The sample size was 425 boys between age 16 and 18. Cluster analysis was more concerned with finding the hierarchical structure of subjects by three dimensional distance of stature. bust girth and sleeve length. The groups forming a partition can be subdivided into 5 and 6 sets by the hierarchical tree of the given subjects. Ward's Minimum Variance Method was applied after extraction of distance matrix by the Standardized Euclidean Distance. All of the above data was analyzed by the computer installed at Korea Advanced Institute of Science and Technology. The major findings, take for instance, of 16 age group can be summarized as follows. The results of cluster analysis of this study: 1. Cluster 1 (32 persons means $18.29\%$ of the total) is characterized with smaller bust girth than that of cluster 5, but stature and sleeve length of the cluster 1 are the largest group. 2. Cluster 2 (18 Persons means $10.29\%$ of the total) is characterized with the group of the smallest stature and sleeve length, but bust girth larger than that of cluster 3. 3. Cluster 3(35persons means $20\%$ of the total) is classified with the smallest group of all the stature, bust girth and sleeve length. 4. Cluster 4(60 persons means $34.29\%$ of the total) is grouped with the same value of sleeve length with the mean value of 16 age group, but the stature and bust girth is smaller than the mean value of this age group. 5. Cluster 5(30 persons means $17.14\%$ of the total) is characterized with smaller stature than that of cluster 1, and with larger bust girth than that of cluster 1, but with the same value of the sleeve length with the mean value of the 16 age group.

• Journal of Periodontal and Implant Science
• /
• 제26권3호
• /
• pp.725-734
• /
• 1996

The present study has been performed to revise the forms of early-onset periodontitis(EOP) into the homogeneous phenotypic subsets by cluster analysis using sets of clinical parameters. Retrospective radiographic interproximal alveolar bone levels were measured from cemento-enamel junctions on patients who have previously been diagnosed as having one of EOP during last 5 years. Mean interproximal bone levels(BL) and mesial bone level(Ratio) of 1st molars relative to mean interproximal bone levels of adjacent teeth(lst and 2nd premolars and canines)were calculated on each patient. Using parameters BL and Ratio(BR group) or BL, Ratio and age(BRA group), cluster analysis was performed to revise EOP patients into homogeneous subsets. At least three or four cluster could be homogeneously formed both in BR or BRA groups with statistically significant differences in parameters used among clusters as evidenced by MANOVA test. It was shown that the greater the BL, the smaller the Ratio was. It was also evident that mean interproximal bone levels were lowest aroud 1st molars and/or incisors regardless of cluster types. The results has provided cluster-based studies for identifying laboratory markers responsible for the development of EOP subsets.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.307-309
• /
• 2007

SVDD의 scale prob1em을 해결하기 위하여, 학습 데이터를 sub-groupings하여 group 단위로 SVDD를 통해 학습함으로써 학습 시간을 줄이는, K-means clustering을 이용한 SVDD 방범(KMSVDD)이 제안되었다. 하지만 KMSVDD는 K-means clustering 알고리즘의 본질상 최적의 K값을 정하기 힘들다는 문제와, 동일한 데이터를 학습할지라도 clustered group이 램덤하게 형성되기 때문에 매번 학습의 결과가 달라지는 문제점이 있었다. 또한 데이터의 분포 상태와 관계없이 무조건 타원(dlliptic) 형태의 K개의 cluster로 나누기 때문에 각각의 나눠진 cluster들은 데이터 분포에 대한 특징을 나타내기 힘들게 된다. 이러한 문제점을 해결하기 위하여 본 논문에서는 데이터 분포에서 mode를 먼저 찾은 후 이 mode를 기준으로 clustering하는 Mean Shift clustering 방법을 이용한 SVDD를 제안하고자 한다. 제안된 알고리즘은 KMSVDD와 비교해 데이터 학습 속도에서는 큰 차이가 없으면서도 데이터의 분포 상태를 고려한 형태로 clustering 한 sub-group을 학습하므로 학습의 정확도가 일정하게 되며, 각각의 cluster는 데이터 분표의 특징을 포함하는 효과가 있다. 또한 Mean Shift Kernel의 bandwidth의 결정은 K-Means의 K와는 달리 어느 정도 여유를 갖고 결정되어도 학습 결과에는 차이가 없다. 다양한 데이터들을 이용한 모의실험을 통하여 위의 내용들을 검증하도록 한다.

• 천문학회보
• /
• 제39권2호
• /
• pp.52-52
• /
• 2014

It is widely accepted that the evolution of galaxies is accelerated in dense environments. According to recent studies, however, the evolution by direct interactions between galaxies is known to be most active in a galaxy group rather than in a galaxy cluster. In particular, the central galaxy in a group is closely related to its satellites in the properties such as morphology, color and star formation rate, because those galaxies evolve together in a small-scale environment. Currently, however, it is not yet studied well whether such conformity between bright galaxies and their faint companions remains after a galaxy group falls into a galaxy cluster. Recently, Lee et al. (2014) have found that the colors of bright galaxies show a measurable correlation with the mean colors of faint companions around them in WHL J085910.0+294957, a galaxy cluster at z = 0.3, which may be the vestige of infallen groups in the cluster. As a follow-up study, we study Abell 2744, an ongoing cluster-cluster merger at z = 0.308, using the HST Frontier Fields Survey data. The cluster members are selected based on the distributions of color, size and concentration along magnitude. The correlation in color between bright galaxies and their companions is not found in the full area of Abell 2744. However, when the area is limited to the southeastern part of the Abell 2744 image, the mean color of faint companions shows marginal dependence (> $2{\sigma}$ to Bootstrap uncertainties) on the color of their adjacent bright galaxy. We discuss the implication of these results, focusing on their dependence on local environments.

• 한국과학교육학회지
• /
• 제33권3호
• /
• pp.679-693
• /
• 2013

The purpose of this study is to develop an inventory to classify task commitment types of science learning and to classify highschool students' task commitment types. Firstly, inventory questions were designed following the literature analysis on the task commitment components which involve self confidence, high goal setting, and focused attention. Prototype inventory underwent the content validity test, pilot test, and reliability test. Through these steps, final inventory was input to 462 high school students and underwent the factor analysis and cluster analysis. Factor analysis confirmed three components of task commitment as the three factors of inventory questions. In order to find how many clusters exist, factors of developed inventory became new variables. Each factor's factor mean was calculated and served as the new variable of the cluster analysis. Cluster analysis extracted five clusters as task commitment types. The 5 clusters were suggested by the agglomarative schedule and dendrogram gained from a hierarchical cluster analysis with the setting of the Ward algorithm and Squared Euclidean distance. Based on the factor mean score, traits of each cluster could be drawn out. Inventory developed by this study is expected to be used to identify student commitment types and assess the effectiveness of task commitment enhancement programs.

• Journal of the Korean Statistical Society
• /
• 제28권2호
• /
• pp.235-251
• /
• 1999

In this paper we examine extremal behavior of a $textsc{k}$th-order stationary Markov chain {X\ulcorner} by considering excesses over a high level which typically appear in clusters. Excesses over a high level within a cluster define a cluster damage, i.e., a normalized sum of all excesses within a cluster, and all excesses define a damage point process. Under some distributional assumptions for {X\ulcorner}, we prove convergence in distribution of the cluster damage and obtain a representation for the limiting cluster damage distribution which is well suited for simulation. We also derive formulas for the mean and the variance of the limiting cluster damage distribution. These results guarantee a compound Poisson limit for the damage point process, provided that it is strongly mixing.

• 천문학회지
• /
• 제29권2호
• /
• pp.181-194
• /
• 1996

We present a study of the metallicity of the old open cluster NGC 1245 , based on the Washington CCD photometry obtained using the 0.6 m telescope at the Sobaeksan Observatory, Korea. NGC 1245 has been known to be a unique cluster among the known open clusters in the sense that the previous metallicity estimates for this cluster are much larger $(by\;\sigma)$ than the value expected from the radial metallicity gradient of the old open clusters in Our galaxy. We have estimated the metallicity of the cluster red giants using the four color-color diagrams, obtaining a value for the mean metallicity of $[Fe/H] = -0.04\pm0.05$ dex. The total error including the error of the metallicity calibration, 0.15 dex, is 0.16 dex. The metallicity estimate of NGC 1245 we have obtained in this study is smaller than previous estimates, and is consistent with the radial metallicity gradient of the old open clusters, showing that the mean metallicity of NGC 1245 is not abnormally high. The reddening, distance, and age of the cluster have also been derived using the isochrones based on the convective overshooting models: the reddening $E(B-V) = 0.28\pm0.03$; the distance $d = 2.5\pm0.2 kpc$ (the corresponding galactocentric distance is RGC = 10.7 kpc, and the distance from the galactic plane is z = -0.4 kpc); and the age $t = 1.1\pm0.1 Gyrs$.