• 응용통계연구
• /
• 제20권1호
• /
• pp.23-38
• /
• 2007

세분화는 크게 목적변수의 유무에 따라 분석방법이 달라지게 된다. 본 논문은 특수법인의 경영지표를 바탕으로 목적변수가 존재하지 않을 경우의 세분화를 통해 전략방향을 도출하는 사례 연구를 제안하고자 한다. 군집분석을 통한 세분화의 경우, 많은 변수를 사용하여 분류를 하게 되면 군집별 특성화가 어렵게 된다. 따라서 군집의 특성을 잘 반영할 수 있는 대표적 요인변수를 요인분석을 통해 추출하고 이 대표요인을 바탕으로 2단계 군집분석을 통한 세분화를 고려하였다. 이를 통해 총 6개의 세분화 군집을 도출하고 각 군집 별 강점요인을 강화하고 약점요인을 보완하는 방향으로 전략방향을 설정하여 제안하고자 한다.

• 한국지역사회생활과학회지
• /
• 제21권4호
• /
• pp.619-631
• /
• 2010

The purpose of this study is to segment and examine urban farmers behavior by applying a two-step cluster analysis and multi-nominal logit model. The data were collected by a telephone survey with two-staged stratified random sampling in the cities around the country for the purpose of acquiring representative data. Respondents were asked to describe their awareness of urban agriculture, their agricultural activity, and sociodemographic characteristics. Among 2,000 cases, 381 cases(19.1%) which were of participants in urban agriculture were analysed in SPSS. From the findings, 27.3% of respondents had heard the word 'urban agriculture', and 25.5% of them regarded themselves as urban farmers. Four different clusters were derived from two-step clusters based on motive, place, companion, area and hours. They were 'Large scale hobby farming(cluster 1)', ‘Weekend farm/ hobby farming(cluster 2)', 'Land/ Self-supporting farming(cluster 3)', and 'Small scale hobby farming(cluster 4)'. The result of multinomial logistic regression showed that there were significant differences among these four segmented groups in terms of age, city size and housing type. In other words, there is quite a possibility that urbanites select different urban farming types according to their socio-demographic profiles. Therefore, the urbanite profiles can be used as the basis for promoting policy of several urban agriculture types. According to the result, policy directions for facilitating urban agriculture were presented.

• 국토계획
• /
• 제54권1호
• /
• pp.40-51
• /
• 2019

The ultimate aim of this study is to diagnose the process stage and look at the step change of transition process to see how the step changes. Therefore, in this study, cluster analysis was conducted by examining four types of commercial characteristics such as Retail Homogeneity, Share of Neighbourhood store, Share of chain store, and Share of cafe & Western food store. Through the cluster analysis, three types have been identified. Type1 is the first step which can explain the time before gentrification occurs and when the ratio of neighborhood facilities is the highest. Type2 is the second step that can explain boutique stage where the gentrification occurs. At this time, the ratio of Cafes & Western food restaurant increased and the proportion of neighborhood shops decreased. And Type3, third step is when the mature gentrification occurs. In the analysis of the transition period, it is necessary to monitor the change of the industry in the period from the first stage to the second stage. In the transition period from the second stage to the third stage, It is necessary to constantly monitor such factors as the increase of shops.

• Journal of the Korean Data and Information Science Society
• /
• 제19권2호
• /
• pp.451-461
• /
• 2008

In this study, factor and cluster analysis have been conducted to group the differences in Korean view on national economic policy in the sample of the 2006 Korean General Social Survey (KGSS). According to the 2006 KGSS, the 6 items with a 5-point Likert scale include the questions about whether or the extent to which each respondent supports the specific types of governmental economic policy. In our study, at first, the factor analysis has converted the original 6 items into the 3 composite variables that account for 81% in the total variability. As the second step of factor analysis, factor scores have been computed. Then, the K-means cluster analysis based on the factor scores has been conducted to group the survey respondents into the 3 clusters. In particular, the cross-tabulation analysis has shown that the distribution of the 3 clusters varies with the respondents' socio-demographic characteristics.

• 한국표면공학회:학술대회논문집
• /
• 한국표면공학회 2016년도 추계학술대회 논문집
• /
• pp.111-111
• /
• 2016

Two step PEO ceramic coatings were formed on AZ91 magnesium alloy in $ZrO_2$ nanoparticles and $K_2ZrF_6$ based colloidal electrolyte solution for various voltages. Surface and layers tructure of the coatings was analyzed using SEM (ScanningElectronMicroscope). Structure analysis revealed that surface of the coating was transferred from individual pancake or craters-based structure to cluster-based structure with increasing the voltage of the secondary step process. Further, it was confirmed that the cluster zone was richin Zr-based complexes and formed due to high intensives parks. Increase in the Zr contents as discovered from the EDS analysis confirmed the rise in amorphous form of the Zr-based species, which justified the results of XRD where no increase in the intensity of Zr-based species was observed with increase in voltage. Potentiodynamic polarizariotion and impedance spectroscopy techniques were used to evaluate the corrosion performance of the coatings. The highest corrosion resistance was found for coatings prepared at 240V. The same specimen was found having highest and uniform vickers hardness ~1070.5 HV. The superior mechanical and electrochemical properties of the said coating can be attributed to the defect-less microstructure and the optimal role of $ZrO_2$ nanoparticles in the secondary PEO process at 240V.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 ITC-CSCC -1
• /
• pp.450-453
• /
• 2000

In this paper, an automated segmentation algorithm is proposed for MR brain images using T1-weighted, T2-weighted, and PD images complementarily. The proposed segmentation algorithm is composed of 3 steps. In the first step, cerebrum images are extracted by putting a cerebrum mask upon the three input images. In the second step, outstanding clusters that represent inner tissues of the cerebrum are chosen among 3-dimensional (3D) clusters. 3D clusters are determined by intersecting densely distributed parts of 2D histogram in the 3D space formed with three optimal scale images. Optimal scale image best describes the shape of densely distributed parts of pixels in 2D histogram. In the final step, cerebrum images are segmented using FCM algorithm with it’s initial centroid value as the outstanding cluster’s centroid value. The proposed segmentation algorithm complements the defect of FCM algorithm, being influenced upon initial centroid, by calculating cluster’s centroid accurately And also can get better segmentation results from the proposed segmentation algorithm with multi spectral analysis than the results of single spectral analysis.

• 동아시아경상학회지
• /
• 제7권2호
• /
• pp.21-30
• /
• 2019

Purpose - In this work, we categorize the 21 shopping items which foreign tourists purchase in South Korea and monitor the level of dissimilarity (or similarity) between each item by utilizing distance matrix, and both hierarchical and k-means cluster analyses, respectively, based on several purpose of visit attributes in 2017. In addition, multidimensional scaling (MDS) method is applied for mining visual appearance of proximities among shopping items based on purpose of visit attributes. Research design and methodology - This study is carried out in 2017 by Ministry of Culture, Sports and Tourism and conduct a face-to-face survey of foreign tourists from 20 countries who purchase shopping items in South Korea. CLUSTER, PROXIMITIES and ALSCAL modules in IBM SPSS 23.0 are used to perform this work. Results - We ascertain that 21 shopping items can be classified into five similar groups which have homogeneous traits by going through two-step cluster analysis. We can position homogeneous places of cluster and shopping items joining each cluster. Conclusions - We can relatively assess patterns and characteristics of each shopping item, come by useful information in activating shopping tour based on the actual state of recognition of foreign tourists and practically apply to each tourism industry on underlying results.

• Journal of the Korean Data and Information Science Society
• /
• 제20권3호
• /
• pp.551-562
• /
• 2009

수협의 어선조업정보는 국가 공식 통계가 가지고 있지 못한 위치별 조업정보를 가지고 있다는 장점이 있다. 위치별 조업정보는 해당 지역의 어업피해보상, 자원가치 산출 등을 추정할 수 있어 국가통계자료로의 가치가 매우 높으나 어업인들의 자기 정보의 노출에 대한 기피로 인해 신뢰성이 떨어지는 단점을 지니고 있다. 본 연구는 유용성분석을 통해 이러한 수협의 어선조업정보의 활용을 위한 방안을 제시하고 위치별 조업패턴의 특성을 분류하여 해구별 유사성의 정보를 산출을 목적으로 하였다. 분석결과 수협의 어선조업정보는 정부 생산통계대비 어획량의 약 33% 수준이나 유용성 분석에서 그 패턴과 상관관계가 밀접해 위치별 패턴파악에 유용한 것으로 나타났다. 이를 바탕으로 대해구별 2단계 군집분석을 통해 어획량, 조업일수, 조업척수에 대해 각각 최적의 군집을 구분하고 이를 종합하여 8개의 군집으로 패턴을 구분하였다.

• 한국지능시스템학회논문지
• /
• 제10권4호
• /
• pp.285-292
• /
• 2000

자료 분석 과정을 살펴 보면 1) 자료가 갖는 경향 평가, 2) 클러스터 분석, 3) 클러스터의 타당성 조사라는 과정을 거쳐 이루어진다. 이 분석법은 2) 및 3) 단계의 반복 수행으로 인하여 많은 계산 시간이 소요되므로 비효율적인 방법이라 할 수 있다. 본 논문에서는, 이와 같은 단점을 보완하기 위하여 자료가 갖는 개략적 특성을 파악하여 자료 속에 존재하는 클러스터의 근사적 개수 및 중심을 정한 후, 이 정보를 기존의 일반적인 퍼지 클러스터링 알고리즘에 입력하여 클러스터링을 수행하는 밀도함수를 이용한 계층적 구조의 근사적 클러스터링 알고리즘을 제안하고, 예제를 통하여 제안된 알고리즘의 타당성을 보인다.

• 천문학회지
• /
• 제37권5호
• /
• pp.349-353
• /
• 2004

Magnetic fields are an important ingredient of galaxy clusters and are indirectly observed on cluster scales as radio haloes and radio relics. One promising method to shed light on the properties of cluster wide magnetic fields is the analysis of Faraday rotation maps of extended extragalactic radio sources. We developed a Fourier analysis for such Faraday rotation maps in order to determine the magnetic power spectra of cluster fields. In an advanced step, here we apply a Bayesian maximum likelihood method to the RM map of the north lobe of Hydra A on the basis of our Fourier analysis and derive the power spectrum of the cluster magnetic field. For Hydra A, we measure a spectral index of -5/3 over at least one order of magnitude implying Kolmogorov type turbulence. We find a dominant scale of about 3 kpc on which the magnetic power is concentrated, since the magnetic autocorrelation length is ${\lambda}_B = 3 {\pm} 0.5\;kpc$. Furthermore, we investigate the influences of the assumption about the sampling volume (described by a window function) on the magnetic power spectrum. The central magnetic field strength was determined to be ${\~}7{\pm}2{\mu}G$ for the most likely geometries.