• 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 the Korean Society for information Management
• /
• v.29 no.2
• /
• pp.45-70
• /
• 2012

There have been many methods and algorithms proposed for multidimensional scaling to mapping the relationships between data objects into low dimensional space. But traditional techniques, such as PROXSCAL or ALSCAL, were found not effective for visualizing the proximities between objects and the structure of clusters of large data sets have more than 50 objects. The CLUSCAL(CLUster-oriented SCALing) technique introduced in this paper differs from them especially in that it uses cluster structure of input data set. The CLUSCAL procedure was tested and evaluated on two data sets, one is 50 authors co-citation data and the other is 85 words co-occurrence data. The results can be regarded as promising the usefulness of CLUSCAL method especially in identifying clusters on MDS maps.

• Journal of Korea Society of Industrial Information Systems
• /
• v.11 no.5
• /
• pp.225-235
• /
• 2006

Not a few nations have tried to copy the success of Silicon Valley since 1980's. Research of information communication industry cluster has emerged in Korea since 1990. In this background this paper thoroughly reviewed scholarly literature related to industry cluster with an attempt to extract dominant factors for cluster formation, and designed a survey instrument to measure the status-quo of cluster formation of software industry in Chungbuk Province Korea, as an example. Based on the results from statistical analysis a strategic scheme was proposed to propel software industry cluster formation in the region.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.12
• /
• pp.2603-2608
• /
• 2009

Document clustering is an important method for document analysis and is used in many different information retrieval applications. This paper proposes a new document clustering model using the clustering method based NMF(non-negative matrix factorization) and refinement of documents in cluster by using coherence of cluster. The proposed method can improve the quality of document clustering because the re-assigned documents in cluster by using coherence of cluster based similarity between documents, the semantic feature matrix and the semantic variable matrix, which is used in document clustering, can represent an inherent structure of document set more well. The experimental results demonstrate appling the proposed method to document clustering methods achieves better performance than documents clustering methods.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06c
• /
• pp.79-82
• /
• 1998

본 연구에서는 연결단어 음성인식 상에서 올바른 참조 패턴을 생성하기 위해 Levelbuilding 알고리즘을 이용하여 인식대상 단어의 표본 집합(훈련패턴 집합)으로부터 참조 패턴을 자동적으로 생성하는 알고리즘을 개발하였다. 본 연구는 분한 K-Mans 훈련방법에 기초하고 있으며, Levelbuilding 알고리즘을 이용하여 훈련패턴으로부터 참조 패턴을 생성하는 것이다. 먼저 초기화 과정에서 훈련 패턴을 그에 포함된 단어 수만큼 등간격 분리하여 분리된 단어들을 소속 Cluster로 분류하고 각 Cluster의 Center들로 초기 참조패턴을 구성한다. 그리고 참조패턴, 제어정보 및 Levelbuilding 알고리즘을 이용하여 각 훈련패턴을 분리하고, 분리된 단어들을 소속 Cluster로 분류하여 단어 Cluster집합을 구성한 후 DTW 및 minimax알고리즘을 이용해 각 Cluster의 Center를 구하여 참조 패턴을 생성한다. 참조패턴 구성에 변화가 없을 때까지 전 단계의 참조패턴과 본 알고리즘을 반복 수행하여 최적의 참조패턴을 생성한다. 본 알고리즘을 이용하여 3개 숫자의 연결단어 집합으로부터 영('0')에서 구('9')까지 숫자음에 대한 참조패턴을 자동 생성하였다. 참조패턴 생성과정에서 가정 중요한 처리인 훈련패턴 분리과정을 분석하기 위하여 각 반복과정에서 분리된 정보를 그래프로 도시화하여 확인하였다.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.28 no.4
• /
• pp.39-60
• /
• 2003

In this study, we evaluate financial performance of 21 domestic life insurers using SAW (simple additive weighting), ELECTREII, cluster analysis respectively, and suggest a hybrid approach of combining cluster analysis and ELECTREII to reclassify the life insurers into more meaningful groups according to their respective financial features. We also perform the sensitivity analysis employing ANOVA and Tukey's test to examine the robustness of ELECTREII, which would be influenced by decision maker's subjective preference parameters. Consequently, it is shown that ELECTREII turns out to be a flexible method providing decision makers with useful ranking Information especially under fuzzy decision making situation with incomparable alternatives, and hence it can serve as a complementary method to overcome the weakness of classical cluster analysis.

• Proceeding of EDISON Challenge
• /
• 2013.04a
• /
• pp.205-209
• /
• 2013

본 실험에서는 탄소 cluster 중에서 fullerene 구조를 가질 수 있는 가장 작은 cluster인 $C_{20}$ cluster에 대해, 기존 연구에서 가장 안정한 것으로 제시된 cage(fullerene), bowl, ring의 3가지 구조와 $Si_{20}$ cluster를 모방한 구조 하나의 안정성을 확인하였다. ab-initio calculation을 지원하는 Edison nanophysics의 LCAODFLab을 이용하여 LDA-CA, GGA-PBE 두 가지 방법으로 계산하였다. 계산 값을 바탕으로 각 구조의 원자화에너지를 비교한 결과 LDA와 GGA 모두 cage, bowl, ring의 순서로 안정하였다. 최적화한 구조에 대하여 구조분석을 진행하였다. 최적화 결과 Bowl은 $C_{5v}$, ring은 $D_{10h}$, cage는 $C_{2h}$ 대칭성을 가지는 구조였으며, LDA, GGA 계산 모두 $C_{20}$ 구조의 spin polarization에는 영향을 받지 않았다.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2007.11a
• /
• pp.203-204
• /
• 2007

This paper investigated the influence on partial discharge distribution of various defects at the model power cable joints interface using K-means clustering. As the result of analyzing discharge number distribution of ${\Phi}-n$ cluster, clusters shifted to $0^{\circ}\;and\;180^{\circ}$ with increasing applying voltage. It was confirmed that discharge quantity and euclidean distance between centroids were increased with applying voltage from the analyzing centroid distribution of ${\Phi}-q$ cluster. The degree of dispersion was increased with calculating standard deviation of ${\Phi}-q$ cluster centroid. The tendency both number of discharge and mean value of ${\Phi}-q$ cluster centroid were some different with defect types.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.36 no.1
• /
• pp.24-35
• /
• 2013

A cell formation approach based on cluster analysis is developed for the configuration of manufacturing cells. Cell formation, which is to group machines and parts into machine cells and the associated part families, is implemented to add the flexibility and efficiency to manufacturing systems. In order to develop an efficient clustering procedure, this paper proposes a cluster analysis-based approach developed by incorporating and modifying two cluster analysis methods, a hierarchical clustering and a non-hierarchical clustering method. The objective of the proposed approach is to minimize intercellular movements and maximize the machine utilization within clusters. The proposed approach is tested on the cell formation problems and is compared with other well-known methodologies available in the literature. The result shows that the proposed approach is efficient enough to yield a good quality solution no matter what the difficulty of data sets is, ill or well-structured.

• Journal of the Korean Applied Science and Technology
• /
• v.39 no.1
• /
• pp.63-73
• /
• 2022

The aim of this study was to cluster according to the bone age of elementary school students in order to analyze the physique, physical fitness, and skeletal maturation of each cluter group and to provide basic data for the balanced development of elementary school students through data analysis. The subjects of this study were 2243 students aged 8 to 13 years, and the skeletal maturation were calculated by applying them to the TW3 method score conversion table after the X-ray films were taken. A total of 2 components in physique were measured using a stadiometer(Hanebio, Korea, 2021) and the Inbody 270(Biospace, Korea, 2019), and a total of 7 components in physical fitness, which included muscular strength(Hand Grip Strength), balance(Bass Stick Test), agility(Plate Tapping), power(Standing Long Jump), flexibility(Sit&Reach), muscular endurance(Sit-Up), and cardiovascular endurance(Shuttle Run) were measured as well. K-Means clustering method, cross-tabulation analysis, and one-way variable analysis(ANOVA) were conducted for data processing using the SPSS PC/Program(Version 26.0) and Bristics Studio Tool, and it was considered significant at the level of p< .05. The results of this study may be summarized as follow. First, as a result of clustering using three components of skeletal maturation: retarded, normal, and advanced, cluster 1(Retarded) showed excellence in muscular strength, balance, and agility. cluster 2(Normal) showed poor flexibility, whereas cluster 3(Advanced) showed excellence in muscular strength. Second, as a result of analyzing the differences in physique according to the clustering of elementary school students by their individual characteristics, cluster 3(Advanced) showed excellence in height, weight, and body fat percentage. Third, as a result of analyzing the differences in physical fitness according to the clustering of elementary school students by their individual characteristics, cluster 3(Advanced) showed excellence in Hand Grip Strength(Left, Right), whereas cluster 1(Retarded) showed excellence in Bass Stick Test, and cluster 3(Advanced) showed excellence in Standing Long Jump.