Cluster Cohesion Research Articles

클러스터의 주요항목 가중치 기반 XML 문서 클러스터링

발달된 인터넷 환경과 데이터 교환 표준 언어로서 확정되고 있는 XML을 기반으로 하여 대량의 웹 문서들이 생산되면서 정보 추출의 대상은 자연스럽게 웹 문서로 이동하게 되었다. 이에 따라 급속히 증가하고 있는 XML 문서에 대한 구조, 통합 및 검색을 위한 연구들이 있다. 이 논문에서는 XML 문서들에 대한 질의 처리, 검색 등을 효율적으로 처리하기 위한 기반으로써 빈발구조 중심의 XML 문서를 클러스터링 하는 방법을 제안한다. 첫째 XML 문서를 트리 구조로 표현하여 분리하고 분리된 구조들을 대상으로 빈발하게 발생하는 구조들을 추출한다. 둘째 각 XML 문서에서 추출된 빈발 구조들을 트랜잭션의 항목으로 취급하여 클러스터링을 수행한다. 클러스터링을 수행할 때 각 클러스터의 생성 및 생성된 전체 클러스터의 응집도를 함께 고려하는 주요항목 가중치를 이용한다. 셋째 기존연구와의 비교 실험을 통해 제안하는 방법의 우수성을 증명한다. As the web document of XML, an exchange language of data in the advanced Internet, is increasing, a target of information retrieval becomes the web documents. Therefore, there we researches on structure, integration and retrieval of XML documents. This paper proposes a clustering method of XML documents based on frequent structures, as a basic research to efficiently process query and retrieval. To do so, first, trees representing XML documents are decomposed and we extract frequent structures from them. Second, we perform clustering considering the weight of large items to adjust cluster creation and cluster cohesion, considering frequent structures as items of transactions. Third, we show the excellence of our method through some experiments which compare which the previous methods.

Metric for Measuring the Effectiveness of Clustering of DNA Microarray Expression

BackgroundThe recent advancement of microarray technology with lower noise and better affordability makes it possible to determine expression of several thousand genes simultaneously. The differentially expressed genes are filtered first and then clustered based on the expression profiles of the genes. A large number of clustering algorithms and distance measuring matrices are proposed in the literature. The popular ones among them include hierarchal clustering and k-means clustering. These algorithms have often used the Euclidian distance or Pearson correlation distance. The biologists or the practitioners are often confused as to which algorithm to use since there is no clear winner among algorithms or among distance measuring metrics. Several validation indices have been proposed in the literature and these are based directly or indirectly on distances; hence a method that uses any of these indices does not relate to any biological features such as biological processes or molecular functions.ResultsIn this paper we have proposed a metric to measure the effectiveness of clustering algorithms of genes by computing inter-cluster cohesiveness and as well as the intra-cluster separation with respect to biological features such as biological processes or molecular functions. We have applied this metric to the clusters on the data set that we have created as part of a larger study to determine the cancer suppressive mechanism of a class of chemicals called retinoids.We have considered hierarchal and k-means clustering with Euclidian and Pearson correlation distances. Our results show that genes of similar expression profiles are more likely to be closely related to biological processes than they are to molecular functions. The findings have been supported by many works in the area of gene clustering.ConclusionThe best clustering algorithm of genes must achieve cohesiveness within a cluster with respect to some biological features, and as well as maximum separation between clusters in terms of the distribution of genes of a behavioral group across clusters. We claim that our proposed metric is novel in this respect and that it provides a measure of both inter and intra cluster cohesiveness. Best of all, computation of the proposed metric is easy and it provides a single quantitative value, which makes comparison of different algorithms easier. The maximum cluster cohesiveness and the maximum intra-cluster separation are indicated by the metric when its value is 0.We have demonstrated the metric by applying it to a data set with gene behavioral groupings such as biological process and molecular functions. The metric can be easily extended to other features of a gene such as DNA binding sites and protein-protein interactions of the gene product, special features of the intron-exon structure, promoter characteristics, etc. The metric can also be used in other domains that use two different parametric spaces; one for clustering and the other one for measuring the effectiveness.

Enhancing the Usage Pattern Mining Performance with Temporal Segmentation of QPop Increment in Digital Libraries

The convergence of next-generation Networks and the emergence of new media systems have made media-rich digital libraries popular in application and research. The discovery of media content objects' usage patterns, where QPop Increment is the characteristic feature under study, is the basis of intelligent data migration scheduling, the very key issue for these systems to manage effectively the massive storage facilities in their backbones. In this paper, a clustering algorithm is established, on the basis of temporal segmentation of QPop Increment, so as to improve the mining performance. We employed the standard C-Means algorithm as the clustering kernel, and carried out the experimental mining process with segmented QPop Increases obtained in actual applications. The results indicated that the improved algorithm is more advantageous than the basic one in important indices such as the clustering cohesion. The experimental study in this paper is based on a Media Assets Library prototype developed for the use of the advertainment movie production project for Olympics 2008, under the support of both the Humanistic Olympics Study Center in Beijing, and China State Administration of Radio, Film and TV.

순차패턴에 기반한 XML 문서 클러스터링

As the use of internet is growing, the amount of information is increasing rapidly and XML that is a standard of the web data has the property of flexibility of data representation. Therefore electronic document systems based on web, such as EDMS (Electronic Document Management System), ebXML (e-business extensible Markup Language), have been adopting XML as the method for exchange and standard of documents. So research on the method which can manage and search structural XML documents in an effective wav is required. In this paper we propose the clustering method based on structural similarity among the many XML documents, using typical structures extracted from each document by sequential pattern mining in pre-clustering process. The proposed algorithm improves the accuracy of clustering by computing cost considering cluster cohesion and inter-cluster similarity.

Genetic Algorithms Applied to Multi-Class Clustering for Gene Expression Data

A hybrid GA (genetic algorithm)-based clustering (HGACLUS) schema, combining merits of the Simulated Annealing, was described for finding an optimal or near-optimal set of medoids. This schema maximized the clustering success by achieving internal cluster cohesion and external cluster isolation. The performance of HGACLUS and other methods was compared by using simulated data and open microarray gene-expression datasets. HGACLUS was generally found to be more accurate and robust than other methods discussed in this paper by the exact validation strategy and the explicit cluster number.

Structures and melting of Cu n ( n=13, 14, 19, 55, 56) clusters

Constant-energy molecular dynamics simulation was utilized to study the most stable geometrical structures, binding energy, melting, phase changes of Cu n ( n=13, 14, 19, 55, 56) clusters. These sizes form shell and near shell structures. The cohesion of clusters is modeled by an embedded-atom potential, which contains many-body atomic interaction terms. Phase space coordinates which are generated along high-energy trajectories are used as the initial configurations (500 independent configurations) for thermal quenching in order to obtain the most stable isomers. The melting temperatures of the clusters are estimated. The melting-like transition is described in terms of relative root-mean-square bond-length fluctuations, specific heats, and caloric curves.

Pair- and many-atom interactions in the cohesion of surface clusters:PdxandIrxon W(110)

In order to explore cohesion in surface clusters, interactions between Pd atoms adsorbed on W(110) have been characterized in observations of individual adatoms using a helium-cooled field ion microscope. The free energies of pair interactions $F(\mathbf{R})$ have been derived from detailed measurements of the probability of finding two atoms on an otherwise empty plane at a vector separation R from each other. For Pd, pair free energies are similar to previous results for Ir atoms: interactions extend over distances larger than 13 \AA{}, and show a strong directional anisotropy. Along the close-packed $[11\ifmmode\bar\else\textasciimacron\fi{}1]$ direction, interactions between two Pd atoms are very close to those for Ir adatoms, even though the binding energy of Ir atoms to the substrate is more than twice that of Pd. With quantitative pair free energies available, the contributions from many-atom effects to the cohesion of clusters have been evaluated by examining the transformation from chain structures to two-dimensional islands, which takes place with both Pd and Ir on W(110). Many-atom effects are parametrized in terms of the trio interactions ${F}_{3\mathrm{lin}}$ in linear and ${F}_{3\mathrm{tri}}$ in triangular configurations. It is found that many-atom effects, rather than amounting to small perturbations of the pair contributions, are the major contributors to the cohesion of two-dimensional clusters for both Pd and Ir on W(110).

Many-atom effects in the cohesion of overlayers: iridium on W(110)

Contributions from many-atom effects to the cohesion of Ir clusters on W(110) have been evaluated by relying on directly measured pair free energies. Strong trio interactions must be invoked to account for the structural change from linear chains to two-dimensional islands which takes place when the number of Ir atoms in a cluster increases above 14. Many-atom effects rather than pair interactions are found to be entirely responsible for the cohesion of 2-D Ir islands; pair contributions are actually repulsive. In Ir chains, pair interactions are, at most, comparable to trio contributions.

Atomic Interactions and the Stability of Surface Clusters

AbstractInteractions dictating the shape and stability of surface islands have been examined for a simple model system, palladium clusters on the W(110) plane. Observations in a field ion microscope of the distribution of two Pd adatoms over the surface yield quantitative values for pair interactions. These are found to be complex, extending over distances longer than 10 A and to vary strongly with the orientation of the pair axis on the (110) surface. Using the measured pair energies it has been possible to infer the magnitude of many-atom effects necessary to account for the equilibrium shape of the clusters. Many-atom interactions turn out to make by far the largest contribution to cluster cohesion; modeling of growth phenomena in terms of nearest-neighbor bonds only is clearly problematic.

Structural and Electronic Aspects of the Phase Transitions of Alkali‐atom Fluids

AbstractWe present a study of the metal‐nonmetal and liquid‐vapor transitions of alkali‐atom fluids which integrates structural and electronic properties. The techniques are extensions of those used to study liquid metals and the phase transitions of pairwise interacting fluids. Structural correlations beyond mean‐field are found to be of special importance due to the large, electronically induced, cohesion of clusters.

The study of macrofungal communities: defining adequate sampling units by means of cluster analysis

A fungus-biotope and a cluster analysis of macrofungal communities were done across a sequence of four forest associations in the Laurentide Mountains of Quebec. Comparison between different scales of sampling indicated that the cohesion and individuality of macrofungal clusters tend to increase from the whole mycoflora of a site toward trophic group and niche group levels. At this latter level, the reliability of defined communities is however highly limited by the low species richness and low constancy of their species. Four sampling units of fungi on the forest floor, were defined: ectomycorrhizal, humicolous, folicolous-muscicolous and lignicolous. This division further contributes to define the natural assemblages of fungi within the biocoenose.

Structure, stability, and surface diffusion of clusters: Ir x on Ir(111)

The properties of clusters adsorbed on a crystal made of the same atoms are of interest in understanding crystal growth. Using the field ion microscope, we have therefore examined iridium clusters containing from 2 to 13 atoms held on the close-packed iridium (111) plane. The arrangement of iridium atoms in the clusters in relation to the binding sites for single atoms on Ir(111) has been determined in detailed mapping experiments. In all the clusters examined, atoms sit in nearest-neighbor sites. As the size of the cluster is increased, the likelihood of finding cluster atoms at bulk sites, at which addition of atoms propagates the normal fcc structure, also increases. From tetramers to Ir 13, compact arrangements maximizing the number of close neighbors are favored. Measurements of the temperature for the onset of dissociation of these clusters reveal an unusual dependence of cluster cohesion upon size: the binding energy increases sharply from dimers to trimers, drops for tetramers, and then rises again. For clusters containing 5 or more atoms, the binding energy is relatively insensitive to size. The magnitude of the cohesive energy is also surprising; the bond energy amounts to only ∼ 1 4 that in a bulk crystal. Diffusion of the clusters over the surface is observed prior to dissociation, and the trends in the barrier to motion are similar to those in dissociation. However, for small clusters the activation energy rises more slowly than in dissociation, and for larger clusters it increases more rapidly. Quantitative determinations of the diffusion characteristics for dimers through pentamers again reveal an anomalously low activation energy for tetramers, but the prefactor is insensitive to cluster size, indicating that the mechanism of diffusion is the same independent of the number of cluster atoms. Detailed observations of cluster motion have been made at low temperatures, to identify the individual atomic events in diffusion. From these it appears that diffusion occurs by a series of single atom jumps. Although surface migration is found to be a reasonably simple process, the cohesive properties of these clusters appear quite sensitive to structure, and are markedly different from expectations based either on bulk properties, or on the behavior of clusters in the gas phase.

On the cohesion of small clusters of alkali metals

Recent results on the cohesion of small neutral and ionised clusters of Na and K are interpreted in a simple Hubbard model, leading to values of transfer and Coulomb integrals of about 0.5 to 1 eV. These values justify the use of a Hartree Fock treatment of electron interactions and are coherent with the fairly compact structures of these aggregates Des resultats recents concernant la cohesion de petits amas neutres et ionises de Na et K sont interpretes dans un modele simple de Hubbard. Ils conduisent a des integrales de transfert et de Coulomb de l'ordre de 0,5 a 1 eV. Ces valeurs justifient l'utilisation d'un traitement de Hartree Fock des interactions electroniques et sont coherents avec les structures assez compactes de ces amas

Macro-regions and the global institutional network, 1950–1980

The article presents an analysis of post-war developments in the international network of intergovernmental organizations (IGOs), and identifies and delineates institutionally cohesive subsystems of states below the world level. It links up with and, for IGOs, tests approaches that stress the increasing salience of (mostly regional) subsystems within world politics. After a brief overview of existing literature and theoretical issues and choices, the growth of the number of organizations, their character, and the level of institutional involvement of states in different parts of the world is described. By means of various clustering procedures clusters of states, according to their institutional links, are identified for different years in the 1950–1980 period. Regionalism prevails within the global institutional network and is getting stronger over time. The cohesion of the different clusters diverges widely. Western Europe, Eastern Europe, Latin America and the Arab Group are the only really tight institutional subsystems of some magnitude and stability. Although existing patterns are remarkably stable through these years, recent developments indicate the possible future emergence of a new institutional cluster of Pacific states, including the USA.

Isolation and morphologic characterization of human ovarian carcinoma cell clusters present in effusions.

Serous, mucinous, endometrioid and clear cell human ovarian carcinoma cells were isolated as multicellular aggregates from patient effusions by filtration on nylon mesh of defined porosity and examined by light microscopy. The cell clusters ranged from compact to loosely adherent groups of cells to spheroids with a central lumen surrounded by a cell monolayer. There was considerable variation in cluster morphology between effusions from different patients as well as within effusion from the same patient. Apparent budding of clusters was observed as well as different stages of cluster growth and development. This was observed for all histologic types studied. Electron microscopy of serous, mucinous and clear cell types showed that cells forming clusters were attached to each other by desmosomes, demonstrating that cluster formation did not result from a nonspecific stickiness of cells. Irregular microvilli were present on the external periphery of the various carcinoma cells and a prominent glycocalyx was present on the surface of mucinous carcinoma cells. Extensive interdigitation of cytoplasmic extensions and extended villi was present in mucinous and serous clusters which appeared to strengthen cluster cohesiveness. Nuclei were irregular with prominent nucleoli frequently present. The cell clusters usually remained intact and viable in culture but generally did not attach to glass or plastic substrata, whereas mesothelial cells and nonactivated histiocytes rapidly attached. When carcinoma cell clusters did attach, they were resistant to detachment by trypsin-EDTA treatment, in contrast to the nonmalignant cells.