• The KIPS Transactions:PartB
• /
• v.11B no.7 s.96
• /
• pp.867-874
• /
• 2004

K-Means algorithm is a non-hierarchical (plat) and reassignment techniques and iterates algorithm steps on the basis of K cluster centroids until the clustering results converge into K clusters. In its nature, K-Means algorithm has characteristics which make different results depending on the initial and new centroids. In this paper, we propose the modified K-Means algorithm which improves the initial and new centroids decision methodologies. By evaluating the performance of two algorithms using the 16 weighting scheme of SMART system, the modified algorithm showed $20{\%}$ better results on recall and F-measure than those of K-Means algorithm, and the document clustering results are quite improved.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.504-504
• /
• 2015

수공구조물을 설계하기 위해서는 설계수문량을 빈도해석을 통해 산정할 수 있다. 빈도해석 중 지점빈도해석을 보완한 지역빈도해석을 적용하기 위해서는 군집분석을 통한 지역구분이 무엇보다 중요하다. 또한 스케일 성질(scaling property)은 강우의 시 공간적 특성을 지속기간별 관측된 강우자료를 이용하여 재현기간에 대한 지속기간의 함수로 강우의 IDF곡선을 제시할 수 있는 방법이다. 따라서 스케일 성질을 통해 군집된 지역에서의 강우자료에 적용하여 스케일 인자(scaling exponent)를 추정한 후 수문학적 동질성을 통계적 특성으로 설명하고자 한다. 본 연구를 수행하기에 앞서 군집 분석은 4개의 군집방법(평균연결법, Ward방법, Two-Step방법, K-means방법)을 적용하였고, 한강유역에 위치한 104개의 강우지점은 4개의 지역으로 구분하는 것이 적절하다고 판단되어 비계층적 방법인 k-means방법을 이용하여 지역을 구분하였다. 본 연구에서는 군집된 결과를 바탕으로 4개의 지역으로 구분된 지역에 포함된 강우지점을 대상으로 스케일 인자를 추정하고 수문학적 동질성을 통계적 방법으로 제시하고자 한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.4
• /
• pp.772-780
• /
• 2008

K-means is one of the simplest unsupervised learning algorithms that solve the clustering problem. However K-means suffers the basic shortcoming: the number of clusters k has to be known in advance. In this paper, we propose extensions of X-means, which can estimate the number of clusters using Bayesian information criterion(BIC). We introduce two different versions of algorithm: modified X-means(MX-means) and generalized X-means(GX-means), which employ one full covariance matrix for one cluster and so can estimate the number of clusters efficiently without severe over-fitting which X-means suffers due to its spherical cluster assumption. The algorithms start with one cluster and try to split a cluster iteratively to maximize the BIC score. The former uses K-means algorithm to find a set of optimal clusters with current k, which makes it simple and fast. However it generates wrongly estimated centers when the clusters are overlapped. The latter uses EM algorithm to estimate the parameters and generates more stable clusters even when the clusters are overlapped. Experiments with synthetic data show that the purposed methods can provide a robust estimate of the number of clusters and cluster parameters compared to other existing top-down algorithms.

• KSBB Journal
• /
• v.16 no.4
• /
• pp.381-388
• /
• 2001

A program for integrated gene expression profile analysis such as hierarchical clustering, K-means, fuzzy c-means, self-organizing map(SOM), principal component analysis(PCA), and singular value decomposition(SVD) was made for DNA chip data anlysis by using Matlab. It also contained the normalization method of gene expression input data. The integrated data anlysis program could be effectively used in DNA chip data analysis and help researchers to get more comprehensive analysis view on gene expression data of their own.

• The Korean Journal of Applied Statistics
• /
• v.22 no.4
• /
• pp.723-733
• /
• 2009

The crucial problems of K-means clustering are deciding the number of clusters and initial centroids of clusters. Hence, the steps of K-means clustering are generally consisted of two-stage clustering procedure. The first stage is to run hierarchical clusters to obtain the number of clusters and cluster centroids and second stage is to run nonhierarchical K-means clustering using the results of first stage. Here we provide automated K-means clustering procedure to be useful to obtain initial centroids of clusters which can also be useful for large data sets, and provide software program implemented using R.

• Journal of Internet Computing and Services
• /
• v.13 no.6
• /
• pp.1-8
• /
• 2012

Clustering method is divided into hierarchical clustering, partitioning clustering, and more. K-Means algorithm is one of partitioning clustering and is adequate to cluster so many documents rapidly and easily. It has disadvantage that the random initial centers cause different result. So, the better choice is to place them as far away as possible from each other. We propose a new method of selecting initial centers in K-Means clustering. This method uses triangle height for initial centers of clusters. After that, the centers are distributed evenly and that result is more accurate than initial cluster centers selected random. It is time-consuming, but can reduce total clustering time by minimizing the number of allocation and recalculation. We can reduce the time spent on total clustering. Compared with the standard algorithm, average consuming time is reduced 38.4%.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.11a
• /
• pp.373-376
• /
• 2002

K-Means 알고리즘은 재배치 기법의 일종으로 K 개의 초기 클러스터중심(centroid)를 중심으로 K 개의 클러스터가 될 때까지 클러스터링을 반복하는 것이다. K-Means 알고리즘은 특성상 초기 클러스터 중심과 새롭게 생성된 클러스터 중심에 따라 클러스터링 결과가 달라진다. 본 논문에서는 K-Means Algorithm 의 초기 클러스터중심 선택 방법과 새로운 클러스터 중심 결정 방법을 개선한 변형 K-Means Algorithm을 제안한다. SMART 시스템에서 제안한 16가지 가중치 계산 방식에 의하여 두 알고리즘의 성능을 평가한 결과 제안한 변형 알고리즘이 재현률과 F-Measure 에서 20%이상 향상된 결과를 얻을 수 있었으며 특정 주제 아래 문서가 할당되는 클러스터링 성능이 우수하였다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.2
• /
• pp.167-172
• /
• 2007

Long gene sequences and their products have been studied by many methods. The use of DNA(Deoxyribonucleic acid) microarray technology has resulted in an enormous amount of data, which has been difficult to analyze using typical research methods. This paper proposes that mass data be analyzed using division clustering with the K-means clustering algorithm. To demonstrate the superiority of the proposed method, it was used to analyze the microarray data from rice DNA. The results were compared to those of the existing K-meansmethod establishing that the proposed method is more useful in spite of the effective reduction of performance time.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.1
• /
• pp.69-72
• /
• 2002

We propose a modified splitting method to obtain an initial codebook, which is used to design a codebook. The principle of the proposed method is that the more representative vectors are assigned to the class, which has the mere member training vectors or a lower squared error. The conventional K-means algorithm and the method provided from reference (5) are used to estimate the performance of the designed codebook. In thin work, the proposed method shows better results than the conventional splitting method in all experiments.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.53 no.10
• /
• pp.69-76
• /
• 2016

In this paper, an efficient image denoising method using non-local means (NL-means) method in the transform domain is proposed. Survey for various image denoising methods has been given, and the performances of the image denoising method using NL-means method have been analyzed. We propose an efficient implementation method for NL-means method by calculating the weights for NL-means method in the DCT and LiftLT transform domain. By using the proposed method, the computational complexity is reduced, and the image denoising performance improves by using the characteristics of images in the tranform domain efficiently. Moreover, the proposed method can be applied efficiently for performing image denoising and image rescaling simultaneously. Extensive computer simulations show that the proposed method shows superior performance to the conventional methods.