High-dimensional unsupervised classification via parsimonious contaminated mixtures. (February 2020)

• Record Type: Journal Article
• Authors: Punzo, Antonio; Blostein, Martin; McNicholas, Paul D.
• Is Part Of: Pattern recognition. Volume 98(2020:Feb.)
• Publication Date: 2020-02
• Subjects: EM algorithm -- Factor analysis -- Mixture models -- Model-based clustering -- Heavy-tailed distributions; 62H30 -- 62H25; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2019.107031
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We propose a robust method for simultaneous unsupervised classification and dimensionality reduction for high-dimensional data. The proposed approach is effective for identifying mild outliers in high-dimensional unsupervised classification problems. The proportion of mild outliers is learned and so does not need to be pre-specified. Abstract: The contaminated Gaussian distribution represents a simple heavy-tailed elliptical generalization of the Gaussian distribution; unlike the often-considered t -distribution, it also allows for automatic detection of mild outlying or "bad" points in the same way that observations are typically assigned to the groups in the finite mixture model context. Starting from this distribution, we propose the contaminated factor analysis model as a method for dimensionality reduction and detection of bad points in higher dimensions. A mixture of contaminated Gaussian factor analyzers (MCGFA) model follows therefrom, and extends the recently proposed mixture of contaminated Gaussian distributions to high-dimensional data. We introduce a family of 32 parsimonious models formed by introducing constraints on the covariance and contamination structures of the general MCGFA model. We outline a variant of the expectation-maximization algorithm for parameter estimation. Various implementation issues are discussed, and the novel family of models is compared to well-established approaches on both simulated and real data.