Title: Mixture-based clustering for the ordered stereotype model
Authors: Daniel Fernandez - Victoria University of Wellington (New Zealand) [presenting]
Richard Arnold - Victoria University of Wellington (New Zealand)
Shirley Pledger - Victoria University of Wellington (New Zealand)
Abstract: Many of the methods that deal with clustering in matrices of data are based on mathematical techniques such as distance-based algorithms or matrix decomposition. In general, it is not possible to use statistical inferences or select the appropriateness of a model via information criteria with these techniques because there is no underlying probability model. Additionally, the use of ordinal data is very common (e.g. Likert or pain scale). Recent research has developed a set of likelihood-based finite mixture models for a data matrix of ordinal data. This approach applies fuzzy clustering via finite mixtures to the ordered stereotype model. Fuzzy allocation of rows, columns, and rows and columns simultaneously (biclustering) to corresponding clusters is obtained by performing a Reversible-Jump MCMC sampler. Examples with ordinal data sets will be shown to illustrate the application of this approach.