Selection of the number of clusters via the bootstrap method

Fang, Yixin; Wang, Junhui

doi:10.1016/j.csda.2011.09.003

Cited by 131 publications

(129 citation statements)

References 24 publications

Supporting

Mentioning

124

Contrasting

Unclassified

Order By: Relevance

“…For this work, cluster performance is determined by the bootstrap method, which minimize cluster instability. 33 Figure 6a shows the cluster instability calculated by the bootstrap method for the HCA-average linkage method with the JSD metric in the Fe-Co-Ni system. When the number of clusters is equal to six, the cluster analysis is the most stable.…”

Section: Resultsmentioning

confidence: 99%

Comparison of dissimilarity measures for cluster analysis of X-ray diffraction data from combinatorial libraries

2017

View full text Add to dashboard Cite

Machine learning techniques have proven invaluable to manage the ever growing volume of materials research data produced as developments continue in high-throughput materials simulation, fabrication, and characterization. In particular, machine learning techniques have been demonstrated for their utility in rapidly and automatically identifying potential composition-phase maps from structural data characterization of composition spread libraries, enabling rapid materials fabrication-structure-property analysis and functional materials discovery. A key issue in development of an automated phase-diagram determination method is the choice of dissimilarity measure, or kernel function. The desired measure reduces the impact of confounding structural data issues on analysis performance. The issues include peak height changes and peak shifting due to lattice constant change as a function of composition. In this work, we investigate the choice of dissimilarity measure in X-ray diffraction-based structure analysis and the choice of measure's performance impact on automatic composition-phase map determination. Nine dissimilarity measures are investigated for their impact in analyzing X-ray diffraction patterns for a Fe-Co-Ni ternary alloy composition spread. The cosine, Pearson correlation coefficient, and Jensen-Shannon divergence measures are shown to provide the best performance in the presence of peak height change and peak shifting (due to lattice constant change) when the magnitude of peak shifting is unknown. With prior knowledge of the maximum peak shifting, dynamic time warping in a normalized constrained mode provides the best performance. This work also serves to demonstrate a strategy for rapid analysis of a large number of X-ray diffraction patterns in general beyond data from combinatorial libraries.

show abstract

Section: Resultsmentioning

confidence: 99%

Comparison of dissimilarity measures for cluster analysis of X-ray diffraction data from combinatorial libraries

2017

View full text Add to dashboard Cite

show abstract

“…The nselectboot function is based on the work of Fang and Wang (2012). T authors focus on the concept of stability as robustness to randomness present in the samp Drawing on the work of Wang (2010), they formulate the concept of stability in the follow way: if one draws samples from the population and applies a selected clustering algorith the results of grouping should not be very different.…”

Section: For Each Pairmentioning

confidence: 99%

Using R Packages for Comparison of Cluster Stability

Rozmus

2017

Folia Oeconomica

View full text Add to dashboard Cite

The stability of clustering methods is the issue that has attracted a considerable amount of attention of researchers in recent years. In this respect, the major question that needs to be answered seems to be to what extent the structure discovered by a particular method is actually present in the data. The literature proposes a number of different ways of measuring stability. The theoretical considerations have led to the development of computer tools for the practical implementation of the proposed ways to study stability. The practical tools are available within several R packages, for example, clv, clValid, fpc, ClusterStability, and pvclust. Due to the hypothesis that cluster stability can be the answer to the question about the right number of groups in clustering, the main aim of this article is to compare the results of the studies on clustering stability conducted with three R packages, i.e.: clv, clValid, and fpc.

show abstract

“…The main reason for using FPC is that it is not necessary to define the number of clusters in advance (before running the clustering algorithm) as in k-means clustering algorithm. FPC computes the number of clusters automatically via bootstrap where several times 2 bootstrap samples are drawn from the data and the number of clusters is chosen by optimizing an instability estimation from these pairs, as explained in [13]. This automatic process is crucial, since we want to analyze several products and not every expert would be able to identify the correct number of clusters without some additional training.…”

Section: Using Clustering To Define Reliable Price Rangementioning

confidence: 99%

Using Clustering and Text Mining to Create a Reference Price Database

Carvalho¹,

Paiva²,

Rocha³

et al. 2014

L&NLM

View full text Add to dashboard Cite

Resumo -Desde 2004, a Controladoria-Geral da União (CGU) tem publicado diversos dados relacionados a gastos do governo brasileiro no Portal da Transparência. Em 2010, a CGU começou a publicar diariamente as despesas diretas do Governo Federal. No entanto, inconsistências que prejudicam a transparência e prestação de contas foram encontradas nessa base de dados. Esse artigo apresenta como a CGU usa técnicas de agrupamento e mineração de texto para recuperar informações essenciais para a melhoria da prestação de contas do governo, incluindo o que foi comprado, o preço pago por item, o preço de referência por produto, etc. Essa análise permitiu que a CGU chegasse a algumas conclusões preliminares que são apresentadas nesse trabalho como forma de ilustrar os resultados da pesquisa. Finalmente, essa informação será eventualmente disponibilizada no Portal da Transparência, pertmitindo que todo cidadão possa ter acesso a informação de quanto o governo está realmente pagando, de uma forma geral, pelos produtos comprados. Dessa forma,é possível melhorar a prestação de contas e a transparência dos gastos públicos não apenas dentro dô ambito da CGU, comoórgão de controle interno, mas também para todos os cidadãos brasileiros, que, no final, pagam a conta.Palavras-chave -preço de referência, agrupamento, mineração de texto, gasto público, prestação de conta, transparência, empenho e compra governamental.Abstract -Since 2004, Brazil's Office of the Comptroller General (CGU) has been publishing several data related to government expenditures in the Transparency Portal. In 2010, CGU started publishing daily every financial statement produced by the Federal Government. Nevertheless, inconsistencies which hinder accountability have been found in this data base. This paper presents how CGU uses clustering and text mining techniques to retrieve essential information for a good accountability, which includes what was bought, the price paid per item, a price reference per product, etc. This analysis has allowed CGU to draw some preliminary conclusions which are presented as a means to illustrate the research results. Finally, this information will eventually be incorporated in the Transparency Portal, allowing every citizen to understand how much the Government is really paying, in general, for products. Thus, improving social control and providing a solid accountability not only to CGU, as an internal control agency, but also to Brazil's citizens who, in the end, are the ones paying the bill.

show abstract

Selection of the number of clusters via the bootstrap method

Cited by 131 publications

References 24 publications

Comparison of dissimilarity measures for cluster analysis of X-ray diffraction data from combinatorial libraries

Comparison of dissimilarity measures for cluster analysis of X-ray diffraction data from combinatorial libraries

Using R Packages for Comparison of Cluster Stability

Using Clustering and Text Mining to Create a Reference Price Database

Contact Info

Product

Resources

About