High-dimensional unsupervised classification via parsimonious contaminated mixtures

作者：

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.

摘要

•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.

论文关键词：EM algorithm,Factor analysis,Mixture models,Model-based clustering,Heavy-tailed distributions

论文评审过程：Received 2 May 2018, Revised 5 July 2019, Accepted 3 September 2019, Available online 5 September 2019, Version of Record 12 September 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.107031