• 한국국방경영분석학회지
• /
• 제6권2호
• /
• pp.89-127
• /
• 1980

The multivariate analysis techniques of cluster analysis are examined in this article. The theory and applications of the techniques and computer software concerning these techniques are discussed and sample jobs are included. A hierarchical cluster analysis algorithm, available in the IMSL software package, is applied to a set of data extracted from a group of subjects for the purpose of partitioning a collection of 26 attributes of a weapon system into six clusters of superattributes. A nonhierarchical clustering procedure were applied to a collection of data of tanks considering of twenty-four observations of ten attributes of tanks. The cluster analysis shows that the tanks cluster somewhat naturally by nationality. The principal componant analysis and the discriminant analysis show that tank weight is the single most important discriminator among nationality although they are not shown in this article because of the space restriction. This is a part of thesis for master's degree in operations research.

• 한국의류학회지
• /
• 제20권2호
• /
• pp.373-389
• /
• 1996

The purpose of this study was to classify the somatotype based on the side view and to analyze the characteristics of each somatotype. The subjects were 201 middle-aged women aged from 35 to 54. Data were collected through anthropometry and photometry and analyzed by factor analysis, cluster analysis, analysis of variance, and discriminant analysis. As the result of factor analysis for the classification of somatotypes, 6 factors which explain 80.8% of variance were extracted from 35 photometric measurement. Using factor scores cluster analysis was carried out and the subjects were classified into 4 cluster Each cluster was classified as straight type, turning over type, bending type and swayback according to its position to the relative plumb line and their side view contour. And 4 somatotypes were analyzed by theirs direct anthropometric and indirect Photometric measurment to represent physical characteristics of each group.

• 복식
• /
• 제53권4호
• /
• pp.1-16
• /
• 2003

The purpose of this study is to investigate the sensibility images of Korean Traditional Chumoni. The detailed methodology of this study is as follows. Selections of stimuli to analyse the sensibility images of Korean Traditional Chumoni were made up of 15 stimuli. The survey has been done for the 15 slide stimuli with semantic differential hi-polar scales which are consist of 23 couples of sensibility words. The subjects were 150 female students majoring in clothing and textile. 150 male students majoring in other department and 150 female students majoring in other department in the twenties between 2001. 3. 30 and 2001. 4. 4. The obtained data were analyzed by factor analysis, cluster analysis. ANOVA. The major finds were as follows. 1. To explain the hierarchy of the sensibility of Korean Traditional Chumoni, two image groups were classified, one is noble and characteristic image the other is splendid and intensive image. Finally it represented noble and splendid image. 2. As result of the factor analysis. 3 factors which are Attraction, Decorativeness, Gravity were found to be constructing factors for the sensibility images of Korean Traditional Chumoni. 3. By cluster analysis, 4 clusters were determined according to Korean Traditional Chumoni. Cluster 1 is splendid. multi-colored and realistic in patteren. Cluster 2 is consist of 'true chumonis' and one-colored. Cluster 3 is modal in pattern. Cluster 4 is simple without any decorations. As to the difference of image of Korean Traditional Chumoni, there were significant differences amang 3 factors by cluster Cluster 1 was found most attractive and grave. Cluster 2 was found most decorative. 4. As to the difference of image of Korean Traditional Chumoni, there were significant differences amang 3 factors by decoration. Gold foil was found most attractive and grave. Embroidery was found most decorative. 5. As to the difference of image of Korean traditional chumoni, there were differences in Decorativeness and Gravity by sex and there were differences in Attraction by major.

• 생물정신의학
• /
• 제25권2호
• /
• pp.38-43
• /
• 2018

Objectives Dopamine dysregulation has been regarded as one of the core pathologies in patients with schizophrenia. Since dopamine synthesis capacity has found to be inconsistent in patients with schizophrenia, current classification of patients based on clinical symptoms cannot reflect the neurochemical heterogeneity of the disease. Here we performed new subtyping of patients with first-episode psychosis (FEP) through biotype-based cluster analysis. We specifically suggested basal ganglia structural changes as a biotype, which deeply involves in the dopaminergic circuit. Methods Forty FEP and 40 demographically matched healthy participants underwent 3T T1 MRI. Whole brain parcellation was conducted, and volumes of total 6 regions of basal ganglia have been extracted as features for cluster analysis. We used K-means clustering, and external validation was conducted with Positive and Negative Syndrome Scale (PANSS). Results K-means clustering divided 40 FEP subjects into 2 clusters. Cluster 1 (n = 25) showed substantial volume decrease in 4 regions of basal ganglia compared to Cluster 2 (n = 15). Cluster 1 showed higher positive scales of PANSS compared with Cluster 2 (F = 2.333, p = 0.025). Compared to healthy controls, Cluster 1 showed smaller volumes in 4 regions, whereas Cluster 2 showed larger volumes in 3 regions. Conclusions Two subgroups have been found by cluster analysis, which showed a distinct difference in volume patterns of basal ganglia structures and positive symptom severity. The result possibly reflects the neurobiological heterogeneity of schizophrenia. Thus, the current study supports the importance of paradigm shift toward biotype-based diagnosis, instead of phenotype, for future precision psychiatry.

• 한국의류학회지
• /
• 제28권9_10호
• /
• pp.1300-1311
• /
• 2004

The aim of this study was to provide the 2 and 3 dimensional statistics requisite in the sizing system and design of gloves. The 64 2-dimensional static measurements were selected to provide information about hands. Participants in the study were 824 adults, aged between 18 and 64. To summarize the information from the measurement values, a Factor Analysis and a Cluster Analysis among multivariate analyses were performed. 3-D scanner was used for visual results of hand shape of each cluster. The results were as follows. Twenty-two items were used for the factor and cluster analysis in order to classify the adult hand shape. The variable quantities that are explained by a total of 3 factors amounted to under 79.37% of the variable quantities. The definition results of the factors related to the hands are as follows: Factor 1 is the horizontal dimension, the thickness of hand factor; Factor 2 is the height of the crotch; and Factor 3 is the vertical dimension of the hand. The adults' group hand was divided into 2 clusters according to a cluster analysis using factor scores. The characteristics according to hand type were as follows: Cluster 1 referred to high horizontal dimensions and thickness, rather small vertical dimensions and crotch height; and Cluster 2 represented the rather smaller horizontal dimensions and thickness but longer hand length than Type 1. To provide specific shape data of each cluster, 3-D scanner measurement was performed. 3-dimensional data base was developed for each cluster type and visual information was provided.

• 한국의류학회지
• /
• 제14권3호
• /
• pp.208-215
• /
• 1990

The purpose of this study is to provide the fundamental data of a dummy design for more suitable ready made clothing by making a pattern of somatic types and analyzing their morphological characteristics in accordance with different pattern of somatic types. The side view silhouettes of 90 junior high school girls of age $13\~16$ in seoul urban area were measured by means of the plan photographing and the low data were examined by principal component analysis, while the principal component analysis was applied and three components were extracted and then interpreted to explain to variation of the form of the body. Using three components respectively the cluster analysis was carried out and the subject classified into 4 cluster The following outcomes are obtained. . The results of principal component analysis of this study would be turned out the three; 1) The first principal component shows the degree of erectness or stoop of the figure. 2) The second principal component was a stature length or a growth rate. 3) The third principal component was the obesity component. 2. The results of cluster analysis by using three principal component analysis would be turned out the four cluser; 1) Cluster 1 ($29\%$ of the total) is characterized with lower stature. 2) Cluster 2 ($21\%$ of the total) is characterized with backward somatotype, and the highest leg. 3) Cluster 3 ($23\%$ of the total) is thicked back of neck. 4) Cluster 4 ($27\%$ of the total) is characterized with forward somatotype, and highest stature, height.

• 동아시아식생활학회지
• /
• 제18권5호
• /
• pp.829-840
• /
• 2008

The principal objectives of this study were : 1) to classify groups via cluster analysis for undergraduate's attributes of restaurant choice, and 2) to assess differences in dining-out behaviors among each restaurant choice cluster. Self-administered questionnaires were completed by 387 students, and the data were analyzed via frequency analysis, chi-square, one-way ANOVA, factor analysis, reliability analysis, cluster analysis and discriminant analysis. Three clusters were obtained by the attributes of restaurant choice from cluster analysis : Cluster 1 "indifferent style", Cluster 2 "ordinary style", and Cluster 3 "highly interested (careful) style". As for connections between the attributes of restaurant choice and distinctive dining-out behaviors in different undergraduates clusters, an indifferent style was rampant among the male students in their lower academic years. This group of students responded that they ate out only to satisfy their hunger, and stayed in a restaurant for one hour or less. Their friends made the choice of a restaurant on their behalf. The students in the "ordinary style" group spent between 5 and 10 thousand won to dine out, and stayed in a restaurant with their friends for two hours or less. They made the choice of a restaurant on their own, or together with their friends. A highly interested (careful) style prevailed among the female students in the upper academic year. This group of students ate out in order to mingle with their friends or colleagues rather than to satisfy their hunger, and they generally made a choice of a restaurant on their behalf.

• 종양간호연구
• /
• 제9권2호
• /
• pp.77-85
• /
• 2009

Purpose: The purpose of this study was to identify symptom cluster experienced by patients with advanced non-small cell lung cancer (NSCLC) on gefitinib treatment. In addition, this study assessed the patterns in severity of the symptom cluster and differences in quality of life (QOL) and function among subgroups by the severity of symptom cluster. Methods: This study was conducted as a secondary analysis of symptoms of 72 patients from a mother study. Factor analysis was used to identify symptom clusters measured with EORTC QLQ-C30 and LC13 symptom related items. Results: Three symptom clusters were identified: cluster 1 was comprised of fatigue, anorexia and dysphagia; cluster 2 of dyspnea, cough and insomnia; and cluster 3 of pain, constipation and nausea/vomiting. These three symptom clusters were improved one week after gefitinib administration. The group with more severe symptom clusters showed significantly lower QOL and function than the group with less severe symptom clusters. Conclusion: Since symptom clusters experienced by the patients with advanced NSCLC influenced on the QOL and function, it is important for nurses to understand and observe their symptom clusters. In addition, there is an necessity to develop nursing interventions to effectively care patients with the symptom clusters.

• 한국식생활문화학회지
• /
• 제23권3호
• /
• pp.377-387
• /
• 2008

The objective of this study was to use cluster analysis to determine differences in eating-out behavior among grouped clusters of female consumers after each cluster was divided based on lifestyle patterns. The data were collected by interview survey from a biased sample of 1,300 females, ranging from ages 20 to 59, and living in residential districts of Seoul. Reliability analysis, factor analysis, cluster analysis, cross-tabulation analysis, and analysis of variance (ANOVA) were applied to the data. Four lifestyle factors were extracted by lower-division and classified as follows: health condition, consuming, food, and housing lifestyles. Based on these four factors, the female consumers were grouped as three clusters: the consuming-individuality type, rational-pursuit type, and conservative-stability type. The eating-out behavior of each cluster was significantly different in terms of frequency of eating-out, eating-out expenditures, restaurant selection criteria, food preferences, and the purpose for eating-out. Since this study surveyed females from ages 20 to 59, age and demographics were the differential factors in determining the various lifestyle types. Thus, to target the consumers who form a target market, the food industry should consider market segmentation that combines demographic factors such as age, income, and marital status.

• 응용통계연구
• /
• 제34권6호
• /
• pp.905-922
• /
• 2021

라벨 없이 진행되는 비지도 학습 중 하나인 군집분석은 자료에 어떤 그룹이 내포되어 있는지 사전 지식이 없을 경우에 군집을 발굴하고, 군집 간의 특성 차이와 군집 안에서의 유사성을 분석하고자 할 때 유용한 방법이다. 기본적인 군집분석 중 하나인 K-means 방법은 변수의 개수가 많아질 때 잘 동작하지 않을 수 있으며, 군집에 대한 해석도 쉽지 않은 문제가 있다. 따라서 고차원 자료의 경우 주성분 분석과 같은 차원 축소 방법을 사용하여 변수의 개수를 줄인 후에 K-means 군집분석을 행하는 Tandem 군집분석이 제안되었다. 하지만 차원 축소 방법을 이용해서 찾아낸 축소 차원이 반드시 군집에 대한 구조를 잘 반영할 것이라는 보장은 없다. 특히 군집의 구조와는 상관없는 변수들의 분산 또는 공분산이 클 때, 주성분 분석을 통한 차원 축소는 오히려 군집의 구조를 가릴 수 있다. 이에 따라 군집분석과 차원 축소를 동시에 진행하는 방법들이 제안되어 왔다. 그 중에서도 본 연구에서는 De Soete와 Carroll (1994)이 제안한 방법론을 확률적인 모형으로 바꿔 군집분석을 진행하는 확률적 reduced K-means를 제안한다. 모의실험 결과 차원 축소를 배제한 군집분석과 Tandem 군집분석보다 더 좋은 군집을 형성함을 알 수 있었고 군집 당 표본 크기에 비해 변수의 개수가 많은 자료에서 기존의 비 확률적 reduced K-means 군집분석에 비해 우수한 성능을 확인했다. 보스턴 자료에서는 다른 군집분석 방법론보다 명확한 군집이 형성됨을 확인했다.