• Communications for Statistical Applications and Methods
• /
• v.22 no.1
• /
• pp.55-67
• /
• 2015

An important problem in cluster analysis is the selection of variables that define cluster structure that also eliminate noisy variables that mask cluster structure; in addition, outlier detection is a fundamental task for cluster analysis. Here we provide an automated K-means clustering process combined with variable selection and outlier identification. The Automated K-means clustering procedure consists of three processes: (i) automatically calculating the cluster number and initial cluster center whenever a new variable is added, (ii) identifying outliers for each cluster depending on used variables, (iii) selecting variables defining cluster structure in a forward manner. To select variables, we applied VS-KM (variable-selection heuristic for K-means clustering) procedure (Brusco and Cradit, 2001). To identify outliers, we used a hybrid approach combining a clustering based approach and distance based approach. Simulation results indicate that the proposed automated K-means clustering procedure is effective to select variables and identify outliers. The implemented R program can be obtained at http://www.knou.ac.kr/~sskim/SVOKmeans.r.

• Journal of applied mathematics & informatics
• /
• v.41 no.4
• /
• pp.861-868
• /
• 2023

Nonlinear network creates complex homotopy structural communication in wireless network medium because of complex distribution approach. Due to this multicast topological connection structure, the queuing probability was non regular principles to create routing structures. To resolve this problem, we propose a Co-cluster homotopy queuing model (Co-CHQT) for Nonlinear Algebraic Topological Structure (NLTS-) for improving poison distribution network communication. Initially this collects the routing propagation based on Nonlinear Distance Theory (NLDT) to estimate the nearest neighbor network nodes undernon linear at x_(a,b)→ax²+bx² = c. Then Quillen Network Decomposition Theorem (QNDT) was applied to sustain the non-regular routing propagation to create cluster path. Each cluster be form with co variance structure based on Two unicast 2(n+1)-Z2(n+1)-Z network. Based on the poison distribution theory X_(a,b) ≠ µ(C), at number of distribution routing strategies weights are estimated based on node response rate. Deriving shorte;'l/st path from behavioral of the node response, Hilbert -Krylov subspace clustering estimates the Cluster Head (CH) to the routing head. This solves the approximation routing strategy from the nonlinear communication depending on Max- equivalence theory (Max-T). This proposed system improves communication to construction topological cluster based on optimized level to produce better performance in distance theory, throughput latency in non-variation delay tolerant.

• Bulletin of the Korean Chemical Society
• /
• v.22 no.1
• /
• pp.103-106
• /
• 2001

• English Language & Literature Teaching
• /
• v.10 no.2
• /
• pp.67-87
• /
• 2004

The purpose of this paper is to show how the coda of a syllable and word-final clusters are represented in the English syllable structure. Previous theories on the syllable assume that there is only one segment in the coda position. And, as we know, the theories that license only one segment in the coda make it difficult to syllabicate the word-final cluster appropriately when more than two segments in the word-final cluster are encountered. I considered three approaches: the previous syllable structure (Selkirk, 1982; Borowsky 1989), sonority sequencing (Giegerich, 1992; Roca, 1999) and feature analysis (Goldsmith, 1990), But, all the considered methods don't give us a satisfactory explanation regarding word-final clusters. Finally, I will suggest a modified syllable representation as an alternative by placing two different appendixes under the Phonological Word which forms a constituent above the syllable node. From this it is possible to explain the former problematic word-final clusters including morphological information asan inflectional suffix in the structure.

• Journal of the Korean Chemical Society
• /
• v.52 no.3
• /
• pp.322-327
• /
• 2008

The geometrical structure, binding energy and vibrational frequencies of HOOCl-H2O cluster were investigated with MP2/6-311G(d,p) and MP2/6-311G(2d,2p) methods. The most stable conformer is skew HOOCl-H2O cluster and the binding energy was 46~48kJ/mol. The trans HOOCl-H2O cluster is less stable than skew form, but the binding energy is big enough to stablize the complex. The vibrational frequencies of skew and trans HOOCl-H2O cluster were calculated and compared with the spectrum of HOOCl.

• Korean Journal of Materials Research
• /
• v.1 no.2
• /
• pp.86-92
• /
• 1991

A potential energy function comprising a two-body potential term which is modified form Morse potential and a three-body potential term which is modified from Axilrod-Teller potential has been developed for small carbon clusters. The structural changes of small carbon clusters $C_2-C_6$ are qualitatively investigated by employing this potential energy function representing the energies of the small carbon cluster isotopes as a function of the three body intensity factor. It is found that the structure of the small carbon cluster changes from open structure to closed one, from complicated structure to simple one, and from three-dimensional structure to two-or-one dimensional one as the degree of the three-body interaction increases.

• Journal of Technology Innovation
• /
• v.15 no.1
• /
• pp.203-233
• /
• 2007

This study investigates how characteristics in the value-chain industrial cluster affect the R&D investment and the output of firms in each industries. For this, we try to extract the inter-industrial networking structure from the input-output table in which 17 manufacturing sectors are included. In particular, we will give shape to the "unit structure" which shows the intermediary flows of goods and services between industries in order to get an unit of final demand in a certain industry. Using this "unit structure", we can try the inter-industrial networking analysis and get some indices of centrality and centralization related to the characteristics of each industries in the value-chain industrial cluster. The results show that the centrality in the value-chain industrial cluster does not have any consequence for the R&D investment and the output in each industries. However, there is a correlation between the centralization in the value-chain industrial cluster and the R&D investment and the output of firms in each industries. These results may be very suggestive in bringing up a new frame of industrial cluster policy in a macro level.

• Journal of ILASS-Korea
• /
• v.13 no.1
• /
• pp.34-41
• /
• 2008

Molecular dynamic simulations have been carried out to study the effect of the nano-structure substrate and its temperature on cluster laminating. The interaction between substrate molecules and liquid molecules was modeled in the molecular scale and simulated by the molecular dynamics method in order to understand behaviors of the liquid cluster on nano-structure substrate. In the present model, the Lennard-Jones potential is applied to mono-atomic molecules of argon as liquid and platinum as nano-structure substrate to perform simulations of molecular dynamics. The effect of wettability on a substrate was investigated for the various beta of Lennard-Jones potential. The behavior of the liquid cluster and nano-structure substrate depends on interface wettability and function of molecules force, such as attraction and repulsion, in the collision progress. Furthermore, nano-structure substrate temperature and beta of Lennard-Jones potential have effect on the accumulation ratio. These results of simulation will be the foundation of coating application technology for micro fabrication manufacturing.

• Journal of KIISE:Information Networking
• /
• v.31 no.1
• /
• pp.62-69
• /
• 2004

The cluster structure reduces the number of retransmission messages, when a broadcast to all hosts in ad hoc network is needed. A cluster maintenance scheme is employed to preserve this advantage from time to time. However, most of the cluster maintenance schemes require additional control messages for cluster reformation as well as control messages for acquiring neighbor information. This mitigates the advantages of employing cluster structure in ad hoc network. In this paper, a cluster maintenance scheme which forces only clusterheads to broadcast control messages during hello time is proposed. When the cluster reformation is needed, the proposed scheme employs a strategy to reduce the control messages to a minimum. In these processes, the proposed scheme doesn't violate the definition of 2-cluster and produces the clusters in fully distributed method. The simulation results prove that our scheme is better than LCC(1).

• Journal of the Ergonomics Society of Korea
• /
• v.29 no.1
• /
• pp.39-46
• /
• 2010

The growth of function in vehicle needs complex display and control system, the In-Vehicle Information System(IVIS). Although current IVISs are widely implemented in commercial vehicles, a new form of IVIS has been recently studied in order to reduce drivers' workloads. The purpose of this study is to suggest an appropriate menu structure of a new type IVIS, to be implemented on the instrument cluster panel, using Taguchi's parameter design. In the research, firstly, functions were selected that are appropriate to control through the instrument cluster among existing functions of current IVISs by quantitative evaluation of ergonomic principles. Then, menu structure alternatives were extracted by investigating priorities to those functions selected. Finally, menu structure alternatives were evaluated through an experiment and suggest the most appropriate one by applying Taguchi's parameter design. Taguchi method was used not only for planning an experiment but also evaluating alternatives. SN ratios were a key value to evaluate the alternatives and to find the most proper one. Through the research, the most appropriate menu structure for the instrument cluster IVIS was finally suggested among the alternatives and it is expected that the results of this research could provide a guideline to the instrument cluster IVIS.