• 정보저장시스템학회:학술대회논문집
• /
• 2005.10a
• /
• pp.83-87
• /
• 2005

Data storage related with writing and retrieving requires high storage capacity, fast transfer rate and less access time in. Today any data storage system can not satisfy these conditions, but holographic data storage system can perform faster data transfer rate because it is a page oriented memory system using volume hologram in writing and retrieving data. System architecture without mechanical actuating pare is possible, so fast data transfer rate and high storage capacity about 1Tb/cm3 can be realized. In this paper, to correct errors of binary data stored in holographic digital data storage system, find cluster centers using clustering algorithm and reduce intensities of pixels around centers. We archive the procedure by two algorithms of C-mean and subtractive clustering, and compare the results of the two algorithms. By using proper clustering algorithm, the intensity profile of data page will be uniform and the better data storage system can be realized.

• ETRI Journal
• /
• v.36 no.3
• /
• pp.333-342
• /
• 2014

Most hyper-ellipsoidal clustering (HEC) approaches use the Mahalanobis distance as a distance metric. It has been proven that HEC, under this condition, cannot be realized since the cost function of partitional clustering is a constant. We demonstrate that HEC with a modified Gaussian kernel metric can be interpreted as a problem of finding condensed ellipsoidal clusters (with respect to the volumes and densities of the clusters) and propose a practical HEC algorithm that is able to efficiently handle clusters that are ellipsoidal in shape and that are of different size and density. We then try to refine the HEC algorithm by utilizing ellipsoids defined on the kernel feature space to deal with more complex-shaped clusters. The proposed methods lead to a significant improvement in the clustering results over K-means algorithm, fuzzy C-means algorithm, GMM-EM algorithm, and HEC algorithm based on minimum-volume ellipsoids using Mahalanobis distance.

• Journal of Computing Science and Engineering
• /
• v.4 no.2
• /
• pp.97-109
• /
• 2010

Nonnegative matrix factorization (NMF) is a popular method for multivariate analysis of nonnegative data, which is to decompose a data matrix into a product of two factor matrices with all entries restricted to be nonnegative. NMF was shown to be useful in a task of clustering (especially document clustering), but in some cases NMF produces the results inappropriate to the clustering problems. In this paper, we present an algorithm for orthogonal nonnegative matrix factorization, where an orthogonality constraint is imposed on the nonnegative decomposition of a term-document matrix. The result of orthogonal NMF can be clearly interpreted for the clustering problems, and also the performance of clustering is usually better than that of the NMF. We develop multiplicative updates directly from true gradient on Stiefel manifold, whereas existing algorithms consider additive orthogonality constraints. Experiments on several different document data sets show our orthogonal NMF algorithms perform better in a task of clustering, compared to the standard NMF and an existing orthogonal NMF.

• ETRI Journal
• /
• v.32 no.3
• /
• pp.351-361
• /
• 2010

Multiple rotation-based transformation (MRBT) was introduced recently for mitigating the apriori-knowledge independent component analysis (AK-ICA) attack on rotation-based transformation (RBT), which is used for privacy-preserving data clustering. MRBT is shown to mitigate the AK-ICA attack but at the expense of data utility by not enabling conventional clustering. In this paper, we extend the MRBT scheme and introduce an augmented rotation-based transformation (ARBT) scheme that utilizes linearity of transformation and that both mitigates the AK-ICA attack and enables conventional clustering on data subsets transformed using the MRBT. In order to demonstrate the computational feasibility aspect of ARBT along with RBT and MRBT, we develop a toolkit and use it to empirically compare the different schemes of privacy-preserving data clustering based on data transformation in terms of their overhead and privacy.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.8 no.3
• /
• pp.196-201
• /
• 2008

Statistical learning theory has three analytical tools which are support vector machine, support vector regression, and support vector clustering for classification, regression, and clustering respectively. In general, their performances are good because they are constructed by convex optimization. But, there are some problems in the methods. One of the problems is the subjective determination of the parameters for kernel function and regularization by the arts of researchers. Also, the results of the learning machines are depended on the selected parameters. In this paper, we propose an efficient method for objective determination of the parameters of support vector clustering which is the clustering method of statistical learning theory. Using evolutionary algorithm and bootstrap method, we select the parameters of kernel function and regularization constant objectively. To verify improved performances of proposed research, we compare our method with established learning algorithms using the data sets form ucr machine learning repository and synthetic data.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.41 no.2
• /
• pp.17-33
• /
• 2016

Clustering is one of the leading methods to analyze big data and is used in many different fields. This study deals with Clustering Validity Index (CVI) to verify the effectiveness of clustering results. We compare the performance of CVIs with business applications of various field. In this study, the used CVIs for comparing performance are DU, CH, DB, SVDU, SVCH, and SVDB. The first three CVIs are well-known ones in the existing research and the last three CVIs are based on support vector data description. It has been verified with outstanding performance and qualified as the application ability of CVIs based on support vector data description.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.8 no.2
• /
• pp.116-120
• /
• 2008

Clustering is a method for unsupervised learning. General clustering tools have been depended on statistical methods and machine learning algorithms. One of the popular clustering algorithms based on machine learning is the self organizing map(SOM). SOM is a neural networks model for clustering. SOM and extended SOM have been used in diverse classification and clustering fields such as data mining. But, SOM has had a problem determining optimal number of clusters. In this paper, we propose an improvement of SOM using gap statistic and probability distribution. The gap statistic was introduced to estimate the number of clusters in a dataset. We use gap statistic for settling the problem of SOM. Also, in our research, weights of feature nodes are updated by probability distribution. After complete updating according to prior and posterior distributions, the weights of SOM have probability distributions for optima clustering. To verify improved performance of our work, we make experiments compared with other learning algorithms using simulation data sets.

• Journal of Computing Science and Engineering
• /
• v.1 no.2
• /
• pp.195-210
• /
• 2007

Dwarf is a highly compressed structure, which compresses the cube by eliminating the semantic redundancies while computing a data cube. Although it has high compression ratio, Dwarf is slower in querying and more difficult in updating due to its structure characteristics. We all know that the original intention of data cube is to speed up the query performance, so we propose two novel clustering methods for query optimization: the recursion clustering method which clusters the nodes in a recursive manner to speed up point queries and the hierarchical clustering method which clusters the nodes of the same dimension to speed up range queries. To facilitate the implementation, we design a partition strategy and a logical clustering mechanism. Experimental results show our methods can effectively improve the query performance on data cubes, and the recursion clustering method is suitable for both point queries and range queries.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2001.01a
• /
• pp.464-469
• /
• 2001

CRM(Customer Relationship Management : CRM) is an advanced marketing supporting system which analyze customers\` transaction data and classify or target customer groups to effectively increase market share and profit. Many engines were developed to implements the function and those for classification and clustering are considered core ones. In this study, an improved clustering method based on SOM(Self-Organizing Maps : SOM) is proposed. The proposed clustering method finds the optimal number of clusters so that the effectiveness of clustering is increased. It considers all the data types existing in CRM data warehouses. In particular, and adaptive algorithm where the concepts of degeneration and fusion are applied to find optimal number of clusters. The feasibility and efficiency of the proposed method are demonstrated through simulation with simplified data of customers.

• The KIPS Transactions:PartB
• /
• v.11B no.4
• /
• pp.485-490
• /
• 2004

In data representation, the clustering performs a grouping process which combines given data into some similar clusters. The various similarity measures have been used in many researches. But, the validity of clustering results is subjective and ambiguous, because of difficulty and shortage about objective criterion of clustering. The fuzzy clustering provides a good method for subjective clustering problems. It performs clustering through the similarity matrix which has fuzzy membership value for assigning each object. In this paper, for objective fuzzy clustering, the clustering algorithm which joins principal components analysis as a dimension reduction model with bayesian learning as a statistical learning theory. For performance evaluation of proposed algorithm, Iris and Glass identification data from UCI Machine Learning repository are used. The experimental results shows a happy outcome of proposed model.