• 대기
• /
• 제17권1호
• /
• pp.45-53
• /
• 2007

In order to investigate the impacts of heat wave on human health, cluster analysis of meteorological elements (e.g., temperature, dewpoint, sea level pressure, visibility, cloud amount, and wind components) for identifying offensive synoptic air masses is employed. Meteorological data at Seoul during the past 30 years are used. The daily death data at Seoul are also employed. Occurrence frequency of heat waves which is defined by daily maximum temperature greater than the threshold temperature (i.e., $31.2^{\circ}C$) was analyzed. The result shows that the frequency and duration of heat waves at Seoul are increasing during the past 30 years. In addition, the increasing trend of the frequency and duration clearly appears in late spring and early autumn as well as summer. Factor analysis shows that 65.1% of the total variance can be explained by 4 components which are linearly independent. Eight clusters (or synoptic air masses) were classified and found to be optimal for representing the summertime air masses at Seoul, Korea. The results exhibit that cluster-mean values of meteorological variables of an offensive air mass (or cluster) are closely correlated with the observed and standardized deaths.

• 복식
• /
• 제36권
• /
• pp.43-52
• /
• 1998

This study was to analyze properties of the foot form through a development figure of the surface area by factor analysis, cluster analysis, and reglation analysis. This study was performed 71 college women between 19 adn 23 years old residing in Pusan urban area. The result was as follows; 1. There are four factors selected by factor analysis. The 1st factor signifies the surface area of the toe. The 2nd factor signifies the surface area between Metatarsal circumfer-ence and Tarsal circumference. The 3rd factor signifies the surface area between Toe circumference and Metatarsal circumference. The 4th factor signifies the surface area between Heel-ankle circumference. 2. There are three clusters selected by fast cluster. Cluster 1 : The foot size, most parts of foot surface area and total foot surface area are relatively small to other clusters, but the surface area between Metatarsal circumference and Tarsal circumference is relatively large to other clusters. Cluster 2 : The foot size, all parts of foot surface area and total foot surface area are relatively large to other clusters as Cluster 1, but the surface area of the sole of the heel and the ankle is relatively large to other clusters. 3. Total surface area of the foot=-274.94+24.01* Foot lengh+9.28*Metatarsal circumference.

• Journal of the Korean Data and Information Science Society
• /
• 제24권4호
• /
• pp.783-791
• /
• 2013

본 논문에서는 대구 경북지역의 24개 4년제 대학교의 대학공시센터에 공시한 대학지표 자료를 사용하였다. 이들 대학지표들 중 재학생 충원률과 건강보험 취업률에 대한 지표를 이용하여 유사 특징을 가지고 있는 대학들을 그룹화 (분류)하여 그룹의 특징을 분석하는데 목적이 있다. SPSS의 계층적 군집분석과 서포트벡터 군집분석을 분석방법으로 활용하여 실험한 결과에서 공통으로 도출할 수 있는 정보를 구하였다.

• 산업융합연구
• /
• 제19권4호
• /
• pp.9-15
• /
• 2021

본 연구는 일반고 학생들의 희망, 그릿, 학교행복의 관계를 알아보고자 희망과 그릿의 군집별 학교행복 차이를 알아보았다. 세종시 ㅂ고 2, 3학년 282명 자료를 상관분석과 군집분석(cluster analysis)을 했다. 연구결과, 희망과 그릿 및 학교행복 간에 긍정적인 상관관계가 있고 희망 주도사고와 그릿 노력지속성 간 상관관계가 높았다. 다음으로, 희망과 그릿의 하위변인은 4개 군집으로 나타났다. 학교행복은 '높은 심리자본' 집단이 가장 높고, '높은 주도사고'와, '높은 경로사고'는 차이가 없으며, '낮은 심리자본' 이 가장 낮았다. 연구결과, 일반고 교육과정에 주도사고와 노력지속성이 반영되어야 함을 확인할 수 있었다. 또한, 앞으로 실제적 측정을 위한 설문지 개발 등을 제안하였다.

• 지능정보연구
• /
• 제7권2호
• /
• pp.35-49
• /
• 2001

인터넷 비즈니스나 전자상거래와 연관되어 고객관계관리(Customer Relationship management :CRM)에 대한 관심이 널리 확산됨으로 해서 군집분석에 대한 관심이 한층 높아졌고, 다양한 군집분석 프로그램이 시장에 소개되어 지고 있다. 그런, 군집분석 프로그램들은 다른 데이터 분석 기법과는 달리 그들의 성능을 측정하기가 매우 힘들다. 본 논문에서는 이미 알려져 있는 군집구조를 지닌 인위적 데이터를 사용하여 다양한 군집분석 프로그램을 평가할 수 있는 하나의 방법론을 제시하고, 그 방법론의 유용성을 보여 주기 위해 현재 많이 사용하고 있는 네 가지의 군집분석 프로그램을 본 논문에서 제시한 방법론을 사용하여 평가하는데 그 주요 목적을 두고 있다. 본 연구에서 두 가지의 반복적 군집분석 프로그램(Convergent Cluster Analysis:CCA, SPSS의 Clementine), 전통적인 단순군집 프로그램(One-Shot Clustering Program: Howard-Harris 프로그램), 그리고 IBM의 데이터 마이닝 기법 중 하나인 데모그래픽 군집분석 프로그램의 성능을 비교한 결과, 군집분석을 위하여 다른 군집분석 방법 보다 좀 더 지능적으로 초기치를 생성한 CCA방법이 가장 우월한 성능을 보여 주었다.

• 대한영양사협회학술지
• /
• 제19권1호
• /
• pp.82-94
• /
• 2013

The purposes of this study were to measure the service mind and customer orientation of employees and to identify the effect of service mind on customer orientation in elementary school foodservices. The questionnaires were distributed to foodservice employees of the 19 elementary schools, but collected from 12 schools in Gwangju, Gyeonggi. The statistical data analysis was completed using SPSS (ver. 18.0) for the independent sample t-test, ANOVA, Cronbach's alpha, principal component analysis, hierarchical & K-means cluster analysis, Pearson' correlation analysis, and multiple regression analysis. Foodservice employees highly rated their service mind (3.94 out of 5 points), especially their perceptions on the importance of service (4.13 points). The effort to provide service was significantly different depending on the serving place (P<0.05). Employees had a high level of customer orientation (4.02 points), which was significantly influenced by age, position, or career (P<0.05), and cook license (P<0.01). As a result of cluster analysis for service mind, employees were divided into two groups: a low-service mind group (cluster 1) and a high-service mind group (cluster 2). Cluster 2 had a significantly higher overall customer orientation than cluster 1 (P<0.001). The pride in providing services (${\beta}$=0.390, P<0.01) and the perception of the importance of services (${\beta}$=0.297, P<0.05) showed a significant and positive effect on customer orientation.

• 한국지하수토양환경학회지:지하수토양환경
• /
• 제27권6호
• /
• pp.22-36
• /
• 2022

Water resources in Jeju Island are dependent virtually entirely on groundwater. For groundwater resources, drought damage can cause environmental and economic losses because it progresses slowly and occurs for a long time in a large area. Therefore, this study quantitatively evaluated groundwater level fluctuations using principal component and cluster analyses for 42 monitoring wells in Jeju Island, and further identified the types of groundwater fluctuations caused by drought. As a result of principal component analysis for the monthly average groundwater level during 2005-2019 and the daily average groundwater level during the dry season, it was found that the first three principal components account for most of the variance 74.5-93.5% of the total data. In the cluster analysis using these three principal components, most of wells belong to Cluster 1, and seasonal characteristics have a significant impact on groundwater fluctuations. However, wells belonging to Cluster 2 with high factor loadings of components 2 and 3 affected by groundwater pumping, tide levels, and nearby surface water are mainly distributed on the west coast. Based on these results, it is expected that groundwater in the western area will be more vulnerable to saltwater intrusion and groundwater depletion caused by drought.

• 농업과학연구
• /
• 제43권1호
• /
• pp.40-51
• /
• 2016

This study analyzed settlement condition changes of Chungcheong region's mountain area during 10 years, from 2001 to 2012. We performed factor analysis and cluster analysis of data from the 'Census for Mountain Area' carried out by the Korea Forest Service in 2001 and 2012. Factors 1, 2, 3, and 4 represent 'the superiority of non-agricultural income', 'the industrialized mountainous area', 'residential conveniences and benefits', and 'the use of forest resources', respectively. Clusters 1, 2, 3, 4, and 5 are categorized as 'non-agricultural income', 'various mass social functions', 'production of forest products', 'industrialized mountainous region', and 'ordinary mountain region', respectively. We suggest that cluster 1 has potential for development and should be promoted as a possible tourist attraction by digging up geographically unique themes. Cluster 2 has great potential for development and needs planned management through the maintenance or expansion of existing infrastructure. Cluster 3 has potential for development with various high value added industries uncovered. Cluster 4 shows vitality as it holds plenty of more job opportunities than other regions. Cluster 5 is deteriorating as a mountainous region because of an aging population, and it urgently demands development. For a decade, 45 of the 60 regions belonged to a single category, which are now differentiated broadly into two types: Firstly, deterioration changes to potential development and Secondly, vitality is differentiated into potential development and deterioration.

• 한국경제지리학회지
• /
• 제15권4호
• /
• pp.515-533
• /
• 2012

여러 국가들에서 산업클러스터 개념은 지역과 국가경제의 발전을 위한 전략으로서 중요한 역할을 하고 있다. 산업클러스터는 특정 지역 내 기업과 관련 기능의 입지를 통한 상호작용과 각 주체 간 지식의 교류를 통한 기업의 혁신촉진과 관련된 개념으로서 기본적으로 지식경영의 개념과 연관돼 있다. 지식교류촉진활동은 산업클러스터에서 협회, 정부지원기관, 리더 기업 등에 의해 이루어지는 지식경영활동 중 가장 중요한 활동으로서 실제로 많은 산업클러스터들에서 이런 목적의 활동이 계획적으로 전개가 되고 있다. 본 연구는 이런 지식교류촉진활동의 성과에 영향을 주는 요인을 파악해 보는데 목적이 있다. 연구 결과 1) 지역, 산업의 지식공유의 문화, 2) 지역 내 리더십 존재, 3) 체계적 촉진 활동, 4) 정보시스템 이용, 5) 인적 접촉 경로 등의 요인이 지식교류촉진활동의 성과에 긍정적 영향을 주는 것으로 확인되었다. 본 연구는 기업 임직원들에 대한 직접 설문조사와 수집된 자료의 통계적 분석을 통해 이루어졌다. 분석을 위해 상관분석, 다중회귀분석, 정준상관분석 등의 통계적 분석기법이 이용되었다.

• Journal of Preventive Medicine and Public Health
• /
• 제30권1호
• /
• pp.181-193
• /
• 1997

The purpose of this study was to identify population subgroups with similar patterns of diet quality, physical activity, alcohol consumption and cigarette smoking of Korean. The cluster analysis was conducted using the data from Korea National Health Survey(KNHS) in 1995, which consisted of 5,805 persons. We identified six health behavior typologies : 32.9% of the sample had a good diet but sedentary activity level(good diet lifestyle), 7.2% had high activity level but less diet quality(fitness lifestyle). Individuals in the passive lifestyle cluster(39.1%) had no active health promoting activities but tended to avoid risk taking health behavior such as cigarette smoking and alcohol drinking. 1.1% of the sample were in a drinking cluster, 17.2% in a smoking cluster and 2.5% had a hedonic lifestyle characterized by heavy drinking and smoking. The other characteristics of these lifestyle clusters could be presented by demographic and socioeconomic factors.