Mixtures of Common Skew-t Factor Analyzers
This work addresses clustering challenges in high-dimensional and skewed datasets, representing an incremental improvement over existing factor analyzer models.
The authors tackled the problem of clustering high-dimensional and skewed data by introducing a mixture of common skew-t factor analyzers model, which demonstrated excellent clustering performance on real and simulated data.
A mixture of common skew-t factor analyzers model is introduced for model-based clustering of high-dimensional data. By assuming common component factor loadings, this model allows clustering to be performed in the presence of a large number of mixture components or when the number of dimensions is too large to be well-modelled by the mixtures of factor analyzers model or a variant thereof. Furthermore, assuming that the component densities follow a skew-t distribution allows robust clustering of skewed data. The alternating expectation-conditional maximization algorithm is employed for parameter estimation. We demonstrate excellent clustering performance when our model is applied to real and simulated data.This paper marks the first time that skewed common factors have been used.