• Communications for Statistical Applications and Methods
• /
• 제10권2호
• /
• pp.387-397
• /
• 2003

Generally, most of researches that need a variable-clustering process use an exploratory factor analysis technique or a divisive hierarchical variable-clustering method based on a correlation matrix. And some researchers apply a object-clustering method to a distance matrix transformed from a correlation matrix, though this approach is known to be improper. On this paper an agglomerative hierarchical variable-clustering method based on a correlation matrix itself is suggested. It is derived from a geometric concept by using variate-spaces and a characterizing variate.

• Journal of the Korean Data and Information Science Society
• /
• 제25권5호
• /
• pp.999-1009
• /
• 2014

계층적 군집 분석은 분석 결과를 덴드로그램으로 쉽게 표시할 수 있어서 방대한 양의 마이크로어레이 자료를 탐색하기에 유용하며, 군집된 결과를 이용하여 생물학적 현상을 이해하는데 도움을 준다. 하지만, 계층적 군집방법은 두 군집간의 절대값 거리만을 고려하여 병합하기 때문에 군집 간의 상대적 비유사성은 설명하지 못하는 단점이 있다. 본 연구에서는 상대적 계층적 군집 방법을 소개하고, 마이크로어레이 자료와 같이 다양한 군집의 모양을 가진 모의실험 자료들과 실제 마이크로어레이 자료를 사용하여 상대적 계층적 군집방법과 기존의 계층적 군집 방법을 비교하였다. 두 계층적 군집 방법의 질적 평가는 오분류율, 동질성, 이질성 지표를 이용하여 수행하였다.

• Communications for Statistical Applications and Methods
• /
• 제31권5호
• /
• pp.601-612
• /
• 2024

A functional clustering analysis is a crucial machine learning technique in functional data analysis. Many functional clustering methods have been developed to enhance clustering performance. Moreover, due to the phase variability between functions, elastic functional clustering methods, such as applying the Fisher-Rao metric, which can manage phase variation during clustering, have been developed to improve model performance. However, aligning functions without considering the phase variation can distort functional information because phase variation can be a natural characteristic of functions. Hence, we propose a state-of-the-art functional hierarchical clustering that can manage phase and amplitude variations of functional data. This approach is based on the phase and amplitude separation method using the norm-preserving time warping of functions. Due to its invariance property, this representation provides robust variability for phase and amplitude components of functions and improves clustering performance compared to conventional functional hierarchical clustering models. We demonstrate this framework using simulated and real data.

• 한국경영과학회지
• /
• 제29권1호
• /
• pp.43-56
• /
• 2004

The purpose of clustering is to partition a set of objects into several clusters based on some appropriate similarity measure. In most cases, clustering is considered without any prior information on the number of clusters or the structure of the given data, which makes clustering is one example of very complicated combinatorial optimization problems. In this paper we propose a general-purpose clustering method that can determine the proper number of clusters as well as efficiently carry out clustering analysis for various types of data. The method is composed of two stages. In the first stage, two different hierarchical clustering methods are used to get a reasonably good clustering result, which is improved In the second stage by ASA(accelerated simulated annealing) algorithm equipped with specially designed perturbation schemes. Extensive experimental results are given to demonstrate the apparent usefulness of our ASA clustering method.

• Journal of Korean Neurosurgical Society
• /
• 제52권5호
• /
• pp.466-471
• /
• 2012

Objective : The aim of this study was to devise an objective clustering method for magnetoencephalography (MEG) interictal spike sources, and to identify the prognostic value of the new clustering method in adult epilepsy patients with cortical dysplasia (CD). Methods : We retrospectively analyzed 25 adult patients with histologically proven CD, who underwent MEG examination and surgical resection for intractable epilepsy. The mean postoperative follow-up period was 3.1 years. A hierarchical clustering method was adopted for MEG interictal spike source clustering. Clustered sources were then tested for their prognostic value toward surgical outcome. Results : Postoperative seizure outcome was Engel class I in 6 (24%), class II in 3 (12%), class III in 12 (48%), and class IV in 4 (16%) patients. With respect to MEG spike clustering, 12 of 25 (48%) patients showed 1 cluster, 2 (8%) showed 2 or more clusters within the same lobe, 10 (40%) showed 2 or more clusters in a different lobe, and 1 (4%) patient had only scattered spikes with no clustering. Patients who showed focal clustering achieved better surgical outcome than distributed cases (p=0.017). Conclusion : This is the first study that introduces an objective method to classify the distribution of MEG interictal spike sources. By using a hierarchical clustering method, we found that the presence of focal clustered spikes predicts a better postoperative outcome in epilepsy patients with CD.

• Wind and Structures
• /
• 제39권1호
• /
• pp.1-14
• /
• 2024

Widespread damages from extreme winds have attracted lots of attentions of the resilience assessment of power distribution systems. With many related environmental parameters as well as numerous power infrastructure components, such as poles and wires, the increased challenge of power asset management before, during and after extreme events have to be addressed to prevent possible cascading failures in the power distribution system. Many extreme winds from weather events, such as hurricanes, generate widespread damages in multiple areas such as the economy, social security, and infrastructure management. The livelihoods of residents in the impaired areas are devastated largely due to the paucity of vital utilities, such as electricity. To address the challenge of power grid asset management, power system clustering is needed to partition a complex power system into several stable clusters to prevent the cascading failure from happening. Traditionally, system clustering uses the Binary Decision Diagram (BDD) to derive the clustering result, which is time-consuming and inefficient. Meanwhile, the previous studies considering the weather hazards did not include any detailed weather-related meteorologic parameters which is not appropriate as the heterogeneity of the parameters could largely affect the system performance. Therefore, a fragility-based network hierarchical spectral clustering method is proposed. In the present paper, the fragility curve and surfaces for a power distribution subsystem are obtained first. The fragility of the subsystem under typical failure mechanisms is calculated as a function of wind speed and pole characteristic dimension (diameter or span length). Secondly, the proposed fragility-based hierarchical spectral clustering method (F-HSC) integrates the physics-based fragility analysis into Hierarchical Spectral Clustering (HSC) technique from graph theory to achieve the clustering result for the power distribution system under extreme weather events. From the results of vulnerability analysis, it could be seen that the system performance after clustering is better than before clustering. With the F-HSC method, the impact of the extreme weather events could be considered with topology to cluster different power distribution systems to prevent the system from experiencing power blackouts.

• 품질경영학회지
• /
• 제18권1호
• /
• pp.48-54
• /
• 1990

The purpose of this study is to design effective working systems which adapt to change in human needs by developing an method which forms into optimal groups using the measure of cohesion. Two main results can be derived from the study as follows : First, the clustering method based on the entropic measure of cohesion is predominant with respect to any other methods proposed in designing the work groups, since this clustering criterion includes symmetrical relations of total work groups and the dissimilarity as well as the similarity relations of predicate value, the clustering method based on this criterion is suitable for designing the new work structure. Second, total work group is clustered as the workers who have the equal predicate value and then clustering results are produced through the combinatorial agglomerative hierarchical clustering method. This clustering method present more economic results than the method that clustering the total work group do.

• 한국멀티미디어학회논문지
• /
• 제23권1호
• /
• pp.1-7
• /
• 2020

Clustering methods essentially take a two-step approach; extracting feature vectors for dimensionality reduction and then employing clustering algorithm on the extracted feature vectors. However, for clustering images, the traditional clustering methods such as stacked auto-encoder based k-means are not effective since they tend to ignore the local information. In this paper, we propose a method first to effectively reduce data dimensionality using convolutional auto-encoder to capture and reflect the local information and then to accurately cluster similar data samples by using a hierarchical clustering approach. The experimental results confirm that the clustering results are improved by using the proposed model in terms of clustering accuracy and normalized mutual information.

• KEPCO Journal on Electric Power and Energy
• /
• 제2권1호
• /
• pp.61-69
• /
• 2016

Understanding load patterns and customer classification is a basic step in analyzing the behavior of electricity consumers. To achieve that, there have been many researches about clustering customers' daily load data. Nowadays, the deployment of advanced metering infrastructure (AMI) and big-data technologies make it easier to study customers' load data. In this paper, we study load clustering from the view point of yearly and daily load pattern. We compare four clustering methods; K-means clustering, hierarchical clustering (average & Ward's method) and DBSCAN (Density-Based Spatial Clustering of Applications with Noise). We also discuss the relationship between clustering results and Korean Standard Industrial Classification that is one of possible labels for customers' load data. We find that hierarchical clustering with Ward's method is suitable for clustering load data and KSIC can be well characterized by daily load pattern, but not quite well by yearly load pattern.

• 한국정보과학회논문지:데이타베이스
• /
• 제29권5호
• /
• pp.367-380
• /
• 2002

최근 공간 데이타 분석, 영상 분석 등과 같은 대용량 데이타를 관리하는 다양한 응용 업무들이 증가함에 따라, 대용량의 데이타베이스를 위한 클러스터링 기법이 많이 연구되고 있다. 그 중에서도 계층 클러스터링 기법은 데이타베이스의 계층 분할을 표현하는 계층 트리를 생성하고 이를 이용하여 효율적인 클러스터링을 수행하는 방법으로서, 지금까지는 주로 트리를 하위 계층으로부터 상위 계층으로 생성해 가는 상향식(bottom-up) 계층 클러스터링 기법들이 연구되었다. 이러한 상향식 클러스터링 방법은 트리를 생성하기 위하여 전체 데이타베이스를 한 번 이상 액세스하여야 할 뿐만 아니라, 하위 계층에서부터 검색을 시작하기 때문에 트리의 많은 부분을 검색하여야 하는 문제점이 있다. 본 논문에서는 대부분의 데이타베이스 응용에서 이미 유지하고 있는 다차원 색인을 이용하여 클러스터링을 수행하는 새로운 하향식(top-down) 계층 클러스터링 기법을 제안한다. 일반적으로 다차원 색인에서는 가까운 객체들이 동일한 (혹은 인접한) 페이지에 저장될 가능성이 큰 클러스터링 성질을 가진다. 이러한 다차원 색인의 클러스터링 성질을 사용하면 각 객체들간의 거리를 일일이 계산하지 않고도 이웃한 객체들을 식별할 수 있다. 우선 객체들의 밀도에 기반하여 클러스터를 정형적으로 정의한다. 이를 위하여, 객체를 포함하는 영역의 밀도를 이용한 영역 대조 분할(region contrast partition) 개념을 사용한다. 또, 클러스터링 알고리즘에서의 빠른 검색을 위하여 분기 한정(branch-and-bound) 알고리즘을 사용하며, 여기서의 한계값(bound)을 제안하고 이의 정확성을 이론적으로 증명한다. 실험 결과, 제안한 방법은 상향식 계층 클러스터링 방법인 BIRCH와 비교하여, 정확성 측면에서 우수하거나 유사한 것으로 나타났으며, 데이타 페이지 액세스 횟수를 데이타베이스 크기에 따라 최고 26～187배까지 감소시킨 것으로 나타났다. 이 같은 결과로 볼 때, 제안한 방법은 대용량 데이타베이스에서의 클러스터링 성능을 크게 향상시키는 기법으로서, 일반 데이타베이스 응용에 실용적으로 적용 가능하다고 판단된다.