• 제목/요약/키워드: Cluster Reduction

검색결과 217건 처리시간 0.031초

Soluble Single-Molecule Magnet: Mn12-stearate.

  • 박치동;정둑영
    • Bulletin of the Korean Chemical Society
    • /
    • 제22권6호
    • /
    • pp.611-615
    • /
    • 2001
  • A new polynuclear complex of manganese stearate has been prepared by substitution of acetate with stearic acid. The stearate ion with long alkyl chain was used to isolate molecular $Mn_{12}$ cluster from each other. The $Mn_{12}$-stearate compound prepared is soluble in most organic solvents and resistant against water catalyzed reduction. The $Mn_{12}$-stearate compound shows similar electrochemical, magnetic properties to the pristine $Mn_{12}$-acetate.

Structure Preserving Dimensionality Reduction : A Fuzzy Logic Approach

  • Nikhil R. Pal;Gautam K. Nandal;Kumar, Eluri-Vijaya
    • 한국지능시스템학회:학술대회논문집
    • /
    • 한국퍼지및지능시스템학회 1998년도 The Third Asian Fuzzy Systems Symposium
    • /
    • pp.426-431
    • /
    • 1998
  • We propose a fuzzy rule based method for structure preserving dimensionality reduction. This method selects a small representative sample and applies Sammon's method to project it. The input data points are then augmented by the corresponding projected(output) data points. The augmented data set thus obtained is clustered with the fuzzy c-means(FCM) clustering algorithm. Each cluster is then translated into a fuzzy rule for projection. Our rule based system is computationally very efficient compared to Sammon's method and is quite effective to project new points, i.e., it has good predictability.

  • PDF

Model-based inverse regression for mixture data

  • Choi, Changhwan;Park, Chongsun
    • Communications for Statistical Applications and Methods
    • /
    • 제24권1호
    • /
    • pp.97-113
    • /
    • 2017
  • This paper proposes a method for sufficient dimension reduction (SDR) of mixture data. We consider mixture data containing more than one component that have distinct central subspaces. We adopt an approach of a model-based sliced inverse regression (MSIR) to the mixture data in a simple and intuitive manner. We employed mixture probabilistic principal component analysis (MPPCA) to estimate each central subspaces and cluster the data points. The results from simulation studies and a real data set show that our method is satisfactory to catch appropriate central spaces and is also robust regardless of the number of slices chosen. Discussions about root selection, estimation accuracy, and classification with initial value issues of MPPCA and its related simulation results are also provided.

산화막중간층에 의한 수직자기기록층의 입자크기 미세화 (Fine Granulation of Recording Layer in Perpendicular Magnetic Recording Media Using Oxide-interlayer)

  • 김경환;공석현
    • 한국표면공학회지
    • /
    • 제37권4호
    • /
    • pp.196-199
    • /
    • 2004
  • Seedlayers with low surface energy which increases the density of nucleation sites in the initial growth region of the recording layer deposited on them was studied to reduce grain size in recording layer. The seedlayer with low surface energy was so effective to attain finer grain in magnetic upper-layers. The Ni-Fe-O intermediate layer with low surface energy was found to be effective in reduction of grain size as well as magnetic cluster size of Co-Cr-Ta-Pt recording layer. Furthermore, the reduction of grain size in Co-Cr-Ta-Pt recording layer on Ni-Fe-O intermediate layer with low surface energy led to decrease the noise level in the high recording density region.

The use of support vector machines in semi-supervised classification

  • Bae, Hyunjoo;Kim, Hyungwoo;Shin, Seung Jun
    • Communications for Statistical Applications and Methods
    • /
    • 제29권2호
    • /
    • pp.193-202
    • /
    • 2022
  • Semi-supervised learning has gained significant attention in recent applications. In this article, we provide a selective overview of popular semi-supervised methods and then propose a simple but effective algorithm for semi-supervised classification using support vector machines (SVM), one of the most popular binary classifiers in a machine learning community. The idea is simple as follows. First, we apply the dimension reduction to the unlabeled observations and cluster them to assign labels on the reduced space. SVM is then employed to the combined set of labeled and unlabeled observations to construct a classification rule. The use of SVM enables us to extend it to the nonlinear counterpart via kernel trick. Our numerical experiments under various scenarios demonstrate that the proposed method is promising in semi-supervised classification.

Study of Optimization of Ground Vehicles Routes Aiming to Reduce Operational Costs and to Contribute to a Sustainable Development with the Reduction of Carbon Dioxide in the Atmosphere

  • Clecio, A.;Thomaz, F.;Hereid, Daniela
    • 융합경영연구
    • /
    • 제4권1호
    • /
    • pp.1-8
    • /
    • 2016
  • The purpose of this paper is to discuss the methodology of optimizing delivery route scheduling using a capacity integer linear programming problem model developed to a previous case study. The methodology suggests a two-stage decision: the first, automatic, where the manager will obtain guidance generated by the solution of the linear programming model, later they could use post-optimization techniques to fine tune to the best operational solution. This study has the goal to reduce the size of service companies' ground transportation fleets, aiming not only to reduce costs and increase competitive advantages but also to lower levels of air pollution and its consequences, traffic and, therefore, the levels of carbon dioxide, allowing for a reduction in envir onmental disasters.

Synthesis, Structure, and Reactivity of the [Fe4S4(SR)4]2- (R = 2-, 3-, and 4-Pyridinemethane) Clusters

  • Kim, Yu-Jin;Han, Jae-Hong
    • Bulletin of the Korean Chemical Society
    • /
    • 제33권1호
    • /
    • pp.48-54
    • /
    • 2012
  • The $[Fe_4S_4]^{2+}$ clusters with 2-, 3-, and 4-pyridinemethanethiolate (S2-Pic, S3-Pic, and S4-Pic, respectively) terminal ligands have been synthesized from the ligand substitution reaction of the $(^nBu_4N)_2[Fe_4S_4Cl_4]$ (I) cluster. The new $(^nBu_4N)_2[Fe_4S_4(SR)_4]$ (R = 2-Pic; II, 3-Pic; III, 4-Pic; IV) clusters were characterized by FTIR and UV-Vis spectroscopy. Cluster II was crystallized in the monoclinic space group C2/c with a = 24.530 (5) $\AA$, b = 24.636(4) $\AA$, c = 21.762(4) $\AA$, ${\beta}=103.253(3)^{\circ}$, and Z = 8. The X-ray structure of II showed two unique 2:2 site-differentiated $[Fe_4S_4]^{2+}$ clusters due to the bidentate-mode coordination by 2-pyridinemethanethiolate ligands. Cluster III was crystallized in the same monoclinic space group C2/c with a = 26.0740(18) $\AA$, b = 23.3195(16) $\AA$, c = 22.3720(15) $\AA$, ${\beta}=100.467(2)^{\circ}$, and Z = 8. The 3-pyridinemethanethiolate ligand of III was coordinated to the $[Fe_4S_4]^{2+}$ core as a terminal mode. Cluster IV with 4-pyridinemethanethiolate ligands was found to have a similar structure to the cluster III. Fully reversible $[Fe_4S_4]^{2+}/[Fe_4S_4]^+$ redox waves were observed from all three clusters by cyclic voltammetry measurement. The electrochemical potentials for the $[Fe_4S_4]^{2+}/[Fe_4S_4]^+$ transition decreased in the order of II, III and IV, and the reduction potential changes by the ligands were explained based on the structural differences among the complexes. The complex III was reacted with sulfonium salt of $[PhMeSCH_2-p-C_6H_4CN](BF_4)$ in MeCN to test possible radical-involving reaction as a functional model of the [$Fe_4S_4$]-SAM (S-adenosylmethionine) cofactor. However, the isolated reaction products of 3-pyridinemethanethiolate-p-cyanobenzylsulfide and thioanisole suggested that the reaction followed an ionic mechanism and the products formed from the terminal ligand attack to the sulfonium.

하둡 클러스터의 네트워크 사용량 감소를 위한 블록 재배치 알고리즘 (A Block Relocation Algorithm for Reducing Network Consumption in Hadoop Cluster)

  • 김준상;김창현;이원주;전창호
    • 한국컴퓨터정보학회논문지
    • /
    • 제19권11호
    • /
    • pp.9-15
    • /
    • 2014
  • 본 논문에서는 하둡 클러스터의 네트워크 사용량 감소를 위한 블록 재배치 알고리즘을 제안한다. 하둡 클러스터의 스케줄러는 사용자들에게 작업을 받아 다중 태스크로 작업을 나누어서 각 노드들에게 할당한다. 이 때 스케줄러는 데이터 지역성(Data locality)을 만족시키는 노드에 우선적으로 태스크를 할당한다. 만약 처리할 데이터(블록)가 없는 노드에 태스크가 할당되면 다른 노드로부터 전송받아 처리한다. 클러스터의 블록들은 사용 빈도가 각각 다르기 때문에 노드 간 작업 부하의 차이가 발생하며 이로 인해 노드 간 데이터 전송이 빈번해진다. 그래서 제안하는 블록 재배치 알고리즘은 하둡 스케줄러의 작업 할당 패턴에 따라 블록들을 균등하게 재배치한다. 결국 노드들의 작업부하는 평준화 되고 처리할 블록이 없는 노드에서 태스크를 처리하는 경우가 감소하기 때문에 클러스터의 네트워크 트래픽이 감소한다. 시뮬레이션으로 제안하는 블록 재배치 정책의 성능평가를 진행했으며 기본 지연 스케줄링으로 작업을 처리한 경우와 비교하여 최대 23.3%의 네트워크 사용량 감소를 보였다.

부산지역의 다양한 환경적 요인의 변화에 따른 대기오염측정망 평가 (The Assessment of Air Quality Monitoring Network Considering the Change of Various Environmental Factors in Busan)

  • 유은철;박옥현
    • 한국대기환경학회지
    • /
    • 제22권4호
    • /
    • pp.405-420
    • /
    • 2006
  • This study was conducted to understand the change of spatial environmental factors including populations, air pollution source and land-use in Busan, during the period of 1995 and 2004. Firstly, the grids (5 km $\times$ 5 km) were divided using the TM coordinates of Busan and the statistical data of populations and land-use were marked on each grid during studying period. Secondly, the SO$_2$, NO$_2$ and O$_3$ concentrations of areas where air quality monitoring station was not established were estimated on the basis of these air pollutants measured at close air quality monitoring station by kriging method. In order to understand spatial change of air pollution and to investigate duplication and reduction of existing stations, semivariogram, correlation and cluster analysis were carried out. This study showed that the population increased in 2004 only on 8 grids compared to in 1995. The spatial change of SO$_2$, NO$_2$ and O$_3$ was investigated by semivariogram in Busan area. As the results of semivariogram, the spatial change of 502 become smaller and simpler, while that of NO2,03 become larger and more complex in 2004 than in 1995, According to the result of correlation and cluster analysis, the reduction of measurement item or the relocation of air quality monitoring station can be needed in the high dense grid area.

Web 2.0 Cluster 기반의 공정 및 성과관리 시스템 구축에 따른 운영방안 제시 (Guideline for Web 2.0 Cluster based Process and Performance Management System)

  • 옹호경;안재규;김대영
    • 한국건설관리학회:학술대회논문집
    • /
    • 한국건설관리학회 2007년도 정기학술발표대회 논문집
    • /
    • pp.899-904
    • /
    • 2007
  • 건설현장에서의 공정관리 기법은 매우 큰 비중을 차지한다. 그러나 이러한 공정관리 및 그 성과를 나타내어줄 수 있는 지표나 평가 내용은 아주 미미한 수준에 있다고 할 수 있다. 특히, 수도권이 아닌 지방공사나 중소업체의 경우에는 현실적 어려움이 큰 것이 사실이다. 따라서 실무자들의 현장 상황에 맞는 유연한 공정관리 및 성과측정 시스템이 필요하게 되었다. 본 연구에서는 지방 중소업체들을 대상으로 문제점을 파악하여 웹 기반을 통해 체계적으로 구축한 린 개념의 공정 및 성과관리 시스템에 대해 중소건설업체가 쉽게 접근하고 사용할 수 있도록 운영방안을 제시하였다. 이는 지방 중소업체의 경쟁력을 확보하고 공기단축과 공사비 절감 등의 가시적인 성과를 도모할 뿐만 아니라 공사과정의 성과측정과 준공이후 관련 자료로써의 활용도 기대할 수 있다.

  • PDF