The mixture method of clustering applied to three-way data

Basford, K. E.; McLachlan, Geoffrey J.

doi:10.1007/bf01908066

Cited by 76 publications

(51 citation statements)

References 24 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, this reduction process can cause a loss of information that leads to prefer the development of ad-hoc clustering techniques that incorporate spatial as well as temporal information. Similar to the classification proposed by Fouedjio (2016) for the clustering of spatial data, existing spatial-time clustering models can be distinguished into the following four different approaches: non-spatial time series clustering based on a spatial dissimilarity measure (Izakian et al, 2013); spatially constrained time series clustering (Hu & Sung, 2006;Coppi et al, 2010;Gao & Yu, 2016); density-based clustering (Ester et al, 1996;Wang et al, 2006;Birant & Kut, 2007;Ienco & Bordogna, 2016;Xie et al, 2016); model-based clustering (Basford & McLachlan, 1985;Viroli, 2011;Torabi, 2014Torabi, , 2016.…”

Section: Introductionmentioning

confidence: 99%

Copula-based fuzzy clustering of spatial time series

2017

View full text Add to dashboard Cite

This paper contributes to the existing literature on the analysis of spatial time series presenting a new clustering algorithm called COFUST, i.e. COpula-based FUzzy clustering algorithm for Spatial Time series. The underlying idea of this algorithm is to perform a fuzzy Partitioning Around Medoids (PAM) clustering using copula-based approach to interpret comovements of time series. This generalisation allows both to extend usual clustering methods for time series based on Pearson's correlation and to capture the uncertainty that arises assigning units to clusters. Furthermore, its flexibility permits to include directly in the algorithm the spatial information. Our approach is presented and discussed using both simulated and real data, highlighting its main advantages.

show abstract

Section: Introductionmentioning

confidence: 99%

Copula-based fuzzy clustering of spatial time series

2017

View full text Add to dashboard Cite

show abstract

“…Effectively, this means that the (G x Ex A) matrix with scores is transformed into a (G x G x E) matrix with distances. Another approach is that of Basford & McLachlan (1985) who considered a clustering of genotypes into groups based on the response in the other two modes, environments and attributes simultaneously. By appropriate specification of the underlying model, the mixture maximum likelihood method of clustering allows the (G x E x A) matrix to be handled directly.…”

Section: Methods Of Analysismentioning

confidence: 99%

“…b Maturity is US maturity group classification or estimated equivalent. Basford & McLachlan (1985) restrict their analyses to yield and protein percentage, while Basford (1982) discussed all six attributes.…”

Section: Methodsmentioning

confidence: 99%

“…Previous analyses of these data have been published, notably using ordination and classification (Mungomery et al, 1974;Shorter et al, 1977), individual differences scaling (Basford, 1982), and three-way clustering (Basford & McLachlan, 1985). The adaptation of the genotypes is, therefore, well known so the use of this data set permits some judgement on the usefullness of this method of analysis.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An investigation of multi-attribute genotype response across environments using three-mode principal component analysis

Kroonenberg

Basford

1989

Euphytica

View full text Add to dashboard Cite

SummaryThe usefulness of three-mode principal component analysis to explore multi-attribute genotype-environment interaction is investigated. The technique provides a general description of the underlying patterns present in the data in terms of interactions of the three quantities (attributes, genotypes, and environments) involved. As an example, data from an Australian experiment on the breeding of soybean lines are treated in depth.

show abstract

“…As maximum likelihood has been shown to be superior to the method of moments for the estimation of finite mixtures (cf. Fryer and Robertson 1972), the likelihood approach for finite normal mixtures has recently become increasingly popular (e.g., Wolfe 1970;Day 1969;Symons 1981;McLachlan 1982;Basford and McLachlan 1985).…”

Section: Introductionmentioning

confidence: 99%

A mixture likelihood approach for generalized linear models

Wedel

DeSarbo

1995

Journal of Classification

253

155

View full text Add to dashboard Cite

A mixture model approach is developed that simultaneously estimates the posterior membership probabilities of observations to a number of unobservable groups or latent classes, and the parameters of a generalized linear model which relates the observations, distributed according to some member of the exponential family, to a set of specified covariates within each Class. We demonstrate how this approach handles many of the existing latent class regression procedures as special cases, as well as a host of other parametric specifications in the exponential family heretofore not mentioned in the latent class literature. As such we generalize the McCullagh and Nelder approach to a latent class framework. The parameters are estimated using maximum likelihood, and an EM algorithm for estimation is provided. A Monte Carlo study of the performance of the algorithm for several distributions is provided, and the model is illustrated in two empirical applications.

show abstract

The mixture method of clustering applied to three-way data

Abstract: Clustering, Mixture maximum likelihood, Three-way data,

Cited by 76 publications

References 24 publications

Copula-based fuzzy clustering of spatial time series

Copula-based fuzzy clustering of spatial time series

An investigation of multi-attribute genotype response across environments using three-mode principal component analysis

A mixture likelihood approach for generalized linear models

Contact Info

Product

Resources

About