Mixtures of Hidden Truncation Hyperbolic Factor Analyzers

TL;DR

This paper introduces the Mixture of Hidden Truncation Hyperbolic Factor Analyzers (MHTHFA), a versatile model unifying various heavy-tailed and skewed cluster models for improved clustering and classification.

Contribution

It proposes a new, general mixture model that unifies multiple non-Gaussian factor analyzers, including a novel hidden truncation hyperbolic factor analysis component.

Findings

Demonstrated effectiveness on real datasets for clustering.

Showed improved semi-supervised classification performance.

Unified various non-Gaussian models into a single framework.

Abstract

The mixture of factor analyzers model was first introduced over 20 years ago and, in the meantime, has been extended to several non-Gaussian analogues. In general, these analogues account for situations with heavy tailed and/or skewed clusters. An approach is introduced that unifies many of these approaches into one very general model: the mixture of hidden truncation hyperbolic factor analyzers (MHTHFA) model. In the process of doing this, a hidden truncation hyperbolic factor analysis model is also introduced. The MHTHFA model is illustrated for clustering as well as semi-supervised classification using two real datasets.