Unifying data units and models in (co-)clustering

Biernacki, Christophe; Lourme, Alexandre

doi:10.1007/s11634-018-0325-2

Cited by 2 publications

(2 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, as noted in Section 4.1, the way the data is encoded can have a strong impact on the resulting co-clustering partition. Although there are ways to address the matter in some cases, as detailed in [37], the user should be aware of it. Additionally, the influence of each kind of feature on the resulting row partitions is to be investigated more deeply in a future work.…”

Section: Discussionmentioning

confidence: 99%

“…Consequently they won't be grouped together in a similar column cluster, whereas a simple switch in the order of the levels could change this and lead to grouping these variables together. Note that this problem is not specific to co-clustering and is also present in clustering [37]. While the user should be aware that the results are conditional on the encoding of levels, this is not an issue addressed in this work.…”

Section: Modeling Nominal Datamentioning

confidence: 99%

See 1 more Smart Citation

Model-based co-clustering for mixed type data

Selosse

Jacques

Biernacki

2020

Computational Statistics & Data Analysis

Self Cite

View full text Add to dashboard Cite

The importance of clustering for creating groups of observations is well known.The emergence of high-dimensional data sets with a huge number of features leads to co-clustering techniques, and several methods have been developed for simultaneously producing groups of observations and features. By grouping the data set into blocks (the crossing of a row-cluster and a column-cluster), these techniques can sometimes better summarize the data set and its inherent structure. The Latent Block Model (LBM) is a well-known method for performing co-clustering. However, recently, contexts with features of different types (here called mixed type data sets) are becoming more common. The LBM is not directly applicable to this kind of data set. Here a natural extension of the usual LBM to the "Multiple Latent Block Model" (MLBM) is proposed in order to handle mixed type data sets. Inference is performed using a Stochastic EMalgorithm that embeds a Gibbs sampler, and allows for missing data situations.A model selection criterion is defined to choose the number of row and column clusters. The method is then applied to both simulated and real data sets.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Modeling Nominal Datamentioning

confidence: 99%

Model-based co-clustering for mixed type data

Selosse

Jacques

Biernacki

2020

Computational Statistics & Data Analysis

Self Cite

View full text Add to dashboard Cite

show abstract

A Variational Approximations-DIC Rubric for Parameter Estimation and Mixture Model Selection Within a Family Setting

Subedi

McNicholas

2020

J Classif

View full text Add to dashboard Cite

Mixture model-based clustering has become an increasingly popular data analysis technique since its introduction fifty years ago, and is now commonly utilized within the family setting. Families of mixture models arise when the component parameters, usually the component covariance matrices, are decomposed and a number of constraints are imposed. Within the family setting, we need to choose the member of the family, i.e., the appropriate covariance structure, in addition to the number of mixture components. To date, the Bayesian information criterion (BIC) has proved most effective for model selection, and the expectation-maximization (EM) algorithm is usually used for parameter estimation. To date, this EM-BIC rubric has monopolized the literature on families of mixture models. We deviate from this rubric, using variational Bayes approximations for parameter estimation and the deviance information criterion for model selection. The variational Bayes approach alleviates some of the computational complexities associated with the EM algorithm by constructing a tight lower bound on the complex marginal likelihood and maximizing this lower bound by minimizing the associated Kullback-Leibler divergence. We use this approach on the most famous family of Gaussian mixture models within the literature and real and simulated data are used to compare our approach to the EM-BIC rubric.

show abstract

Unifying data units and models in (co-)clustering

Cited by 2 publications

References 54 publications

Model-based co-clustering for mixed type data

Model-based co-clustering for mixed type data

A Variational Approximations-DIC Rubric for Parameter Estimation and Mixture Model Selection Within a Family Setting

Contact Info

Product

Resources

About