• The KIPS Transactions:PartB
• /
• v.13B no.4 s.107
• /
• pp.401-408
• /
• 2006

In this paper, we propose a new class of segmentation technique for feature extraction based on the statistical and regional classification at each vertical or horizontal line of digital image data. Data is processed and clustered at each line, different from the point or space process. They are designed to segment gray-scale sectional images using a horizontal and vertical line process due to their statistical and property differences, and to extract the feature. The techniques presented here show efficient results in case of the gray level overlap and not having threshold image. Such images are also not easy to be segmented by the global or local threshold methods. Line pixels inform us the sectionable data, and can be set according to cluster quality due to the differences of histogram and statistical data. The total segmentation on line clusters can be obtained by adaptive extension onto the horizontal axis. Each processed region has its own pixel value, resulting in feature extraction. The advantage and effectiveness of the line-cluster approach are both shown theoretically and demonstrated through the region-segmental carotid artery medical image processing.

• Korean Journal of Environment and Ecology
• /
• v.27 no.1
• /
• pp.99-112
• /
• 2013

This study aims to show the changes of characteristics of vegetation structure for 20 years(1991~2010) in Naejangsan National Park. As a result of analysis of actual vegetation, the mixed community of Quercus variabilis and Quercus serrata was distributed with 56.1%, and Q. variabilis community showed in southern steep slope with 17.6%. Pinus densiflora community(5.8%) was observed on the ridge and Carpinu tschonoskii community distributed in the slope of the valley with 6.6%. Zelkova serrata and Prunus sargentii community were distributed in valley. The classification by TWINSPAN, ordination by DCA considering importance percentage and property of vegetation class were divided into 4 communities, which are community I(P. densiflora-Q. variabilis community), community II(Q. variabilis community), community III(C. tschonoskii community) and community IV(Mixed deciduous broad-leaved trees community). The age of Pinus densiflora was 32years old and Q. serrata was 36 years old in the community I, that of Q. variabilis was 64 years old in the community II, Q. serrata was 46 years old and C. tschonoskii was 45 years old in the community III, and Acer palmatum was 54 years old and Cornus controversa was 47 years old in the community IV. As the result of Shannon's index of species diversity, the community Iwas ranged from 0.9751 to 1.4199, community II was ranged from 1.0765 to 1.3278, community III was ranged from 1.0353 to 1.2881, and community IV was ranged from 1.1412 to 1.3807. The change of vegetation structure analyzed through the comparison with results of studies carried out 20 years ago were natural selection of P. densiflora, expansion of Quercus spp. and increase of C. tschonoskii. Especially, A. palmatum is dominated by Q. variabilis in canopy layer like the result of study 20 years ago. A. palmatum was analysed by 14.6% in the canopy layer of only mixed deciduous broad-leaved trees community. As a result of analysis of habitat property of Q. variabilis and A. palmatum, Q. variabilis was distributed in dry area with the low value of pH, O.M., exchangeable cations and Avail. P, and A. palmatum was located in the wet valley with huge value of nourishment. The tendency of reduction of bio-diversity by Sasa borealis is same as previous study but, the distributed areas were reduced in Naejangsan area.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.977-981
• /
• 2006

본 연구는 오산천 하천환경정비로 인하여 저서성 대형무척추동물의 군집변화에 미치는 영향을 파악하기 위하여 2002년 6월부터 2005년 11월까지 5개지점을 선정하여 조사를 실시하였다. 조사기간 중 출현한 저서성 대형무척추동물은 총 4문 7강 16목 38과 81종이었으며, 년도별로는 2002년에는 총 55종이 출현하였으나, 2004년에는 36종으로 종수가 급감하였다. ESB에 따른 군집의 생태점수는 2003년도에 45.4에서 2004년도에는 21.7로 낮아졌다가 2005년도에는 31.2로 높아졌다. 이는 하천환경정비공사로 일시적인 교란을 일으켜 저서성 대형무척추동물의 군집변화에 영향을 미친 것으로 사료된다. 종조성의 변화로는 상류부의 경우 환경정비공사 완료 후 생태계안정화와 추이대가 복원되면서 1급수 지표종인 플라나리아(Dugesia japonica)와 옆새우(Gammarus sp.)가 출현하여 다양한 군집을 형성하고 있었다. 하류의 경우는 군집의 종조성이 빈약하지만 공사가 마무리 단계에 들어가면서 수서생물의 서식처가 안정화되고 있는 것으로 사료된다. 이처럼 저서성 대형무척추동물의 서식에 영향을 주는 하상의 물리적 구조와 이와 연관된 유기물 퇴적층과 토사 퇴적층에 대한 관리가 필요하며, 이를 통하여 하도특성에 맞는 안정된 군집구조가 형성될 수 있도록 대체서식처를 하천환경정비계획 수립 시 고려하여야 한다.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.1
• /
• pp.57-64
• /
• 2009

In this paper, 30 main cities are clustered based on precipitation, temperature, wind speed, photo period, and humidity. We found that the resulting clusters has strong relationships with geographical locations. These results make sense because, although Korea is a small country, Korean weather is known to have strong locality. The largest number of clusters is found when wind speed is used as an interested variable for clustering and the smallest number of clusters is found when photo period is used. The large number of clusters based on wind speed indicates that wind speed is affected easily by local geography.

• Proceedings of the Korea Technology Innovation Society Conference
• /
• 2017.11a
• /
• pp.989-1018
• /
• 2017

본 연구에서는 내 외부 환경, 혁신활동이 기업성과에 어떠한 영향을 미치는지를 파악하기 위해 탐색적 요인분석(Exploratory Factor Analysis), 군집분석(Cluster Analysis), 구조방정식모형(Structural Equation Modeling)을 이용하여 실증분석을 실시하였다. 탐색적 요인분석을 통해 7개의 요인이 추출하였으며, 추출된 요인을 기반으로 군집분석을 시도하였더니 총 4개의 군집(n=1,022)이 형성되었다. 군집 4개의 대해 구조방정식 모형을 활용하여 실증분석을 한 결과, 기술 경쟁 환경에 민감하며, 혁신적인 성향을 가진 군집1은 자체기술개발만이 기업성과에 긍정적 영향을 미치는 것으로 나타났다. 시장 환경에 민감하며, 내향적인 성향을 가진 군집2는 자체기술개발과 공동연구를 통해서만 기업성과에 긍정적 영향을 미치는 것으로 나타났다. 경쟁적인 환경에 민감하며, 혁신적이고 정부/관련기관과의 협력적 성향을 가진 군집3은 공동연구 그리고 매개변수인 정부지원프로그램 활용을 통해 기업성과에 긍정적 영향을 미치는 것으로 나타났으며, 기술도입은 기업성과에 부정적 영향을 미치는 것으로 나타났다. 개방적이고 외부협력적 성향이 강한 군집4는 자체기술개발과 매개변수인 네트워크 활용 및 정부지원프로그램 활용이 기업성과에 긍정적 영향을 미치는 것으로 나타났다.

• The Korean Journal of Applied Statistics
• /
• v.32 no.1
• /
• pp.149-160
• /
• 2019

The electricity consumption time series data of 'A' University from July 2016 to June 2017 is analyzed via nonparametric functional data clustering since the time series data can be regarded as realization of continuous functions with dependency structure. We use a Bouveyron and Jacques (Advances in Data Analysis and Classification, 5, 4, 281-300, 2011) method based on model-based functional clustering with an FEM algorithm that assumes a Gaussian distribution on functional principal components. Clusterwise analysis is provided with cluster mean functions, densities and cluster profiles.

• The Korean Journal of Applied Statistics
• /
• v.34 no.6
• /
• pp.905-922
• /
• 2021

Cluster analysis is one of unsupervised learning techniques used for discovering clusters when there is no prior knowledge of group membership. K-means, one of the commonly used cluster analysis techniques, may fail when the number of variables becomes large. In such high-dimensional cases, it is common to perform tandem analysis, K-means cluster analysis after reducing the number of variables using dimension reduction methods. However, there is no guarantee that the reduced dimension reveals the cluster structure properly. Principal component analysis may mask the structure of clusters, especially when there are large variances for variables that are not related to cluster structure. To overcome this, techniques that perform dimension reduction and cluster analysis simultaneously have been suggested. This study proposes probabilistic reduced K-means, the transition of reduced K-means (De Soete and Caroll, 1994) into a probabilistic framework. Simulation shows that the proposed method performs better than tandem clustering or clustering without any dimension reduction. When the number of the variables is larger than the number of samples in each cluster, probabilistic reduced K-means show better formation of clusters than non-probabilistic reduced K-means. In the application to a real data set, it revealed similar or better cluster structure compared to other methods.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2006.05a
• /
• pp.37-44
• /
• 2006

본 연구는 모바일기기 사용자들의 특성을 알아보고자 사용자들의 모바일기기 휴대의도에 영향을 미치는 변수들을 기준으로 사용자들을 세분화 하였다. 선행문헌연구를 통해 휴대의도에 영향을 미치는 변수들로 유용성, 즐거움, 불안감과 사회적 영향인 주관적 규범, 가시성, 이미지, 자발성 등을 도출하였다. 자료수집을 위해 서울 및 수도권에 거주하고 휴대폰이나 노트북, PDA등을 소유한 대학생 및 대학원생과 직장인들을 대상으로 설문조사를 실시하였다. 휴대의도의 영향변수 분석결과 불안감과 주관적 규범이 휴대의도에 유의한 양의 영향을 미치는 것으로 검증되었다. 이를 기준변수로 군집분석을 통해 전체 표본을 네 집단으로 세분화 하였다. 세분화 결과 군집 1은 주관적 규범만이 영향을 미치는 집단으로 전체응답의 36.25%를 차지하고 있으며, 군집 2는 불안감만이 영향을 미치는 집단으로 전체 응답의 13.44%를 구성하는 것으로 나타났다. 군집 3은 불안감과 주관적 규범 모두 영향을 미치지 않는 집단으로서 전체 응답에서 19.06%를 차지하고 마지막으로 군집 4는 불안감과 주관적 규범 모두 영향을 미치는 집단으로 전체응답의 31.25%를 구성하는 것으로 나타났다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.18 no.3
• /
• pp.35-46
• /
• 2013

In order to develop an accurate analytical model for domestic electricity demand forecasting, we propose a prediction method of the electric power demand pattern by combining SMO classification techniques and a dimension reduction conceptualized subspace clustering techniques suitable for high-dimensional data cluster analysis. In terms of electricity demand pattern prediction, hourly electricity load patterns and the demographic and geographic characteristics can be analyzed by integrating the wireless load monitoring data as well as sub-regional unit of census information. There are composed of a total of 18 characteristics clusters in the prediction result for the sub-regional demand pattern by using census information and power load of Seoul metropolitan area. The power demand pattern prediction accuracy was approximately 85%.

• Economic Analysis
• /
• v.17 no.2
• /
• pp.27-55
• /
• 2011

Although many theoretical studies have tried to explain the volatility in financial markets using models of herd behavior, there have been few empirical studies on dynamic herding due to the technical difficulty of detecting herd behavior with time-series data. Thus, this paper theoretically extends a continuous beliefs system belonging to an agent based economic model by introducing a term representing agents'mutual dependence into each agent's utility function and derives a SV(stochastic volatility)-type econometric model. From this model the time-varying herding parameters are efficiently estimated by a Markov chain Monte Carlo method. Using monthly data of KOSPI and DOW, this paper provides some empirical evidences for stronger herding in the Korean stock market than in the U.S. stock market, and further stronger herding after the global financial crisis than before it. More interesting finding is that time-varying herd behavior has weak autocorrelation and the global financial crisis may increase its volatility significantly.