A robust model-based clustering based on the geometric median and the Median Covariation Matrix

TL;DR

This paper introduces a robust model-based clustering method using median and median covariation matrix to improve Gaussian Mixture Models, implemented in an R package.

Contribution

It proposes a novel robust modification of the EM algorithm for clustering, replacing mean and variance estimates with median-based estimators.

Findings

Enhanced robustness against outliers in clustering results

Implementation available in R package RGMM on CRAN

Improved clustering stability in contaminated data scenarios

Abstract

Grouping observations into homogeneous groups is a recurrent task in statistical data analysis. We consider Gaussian Mixture Models, which are the most famous parametric model-based clustering method. We propose a new robust approach for model-based clustering, which consists in a modification of the EM algorithm (more specifically, the M-step) by replacing the estimates of the mean and the variance by robust versions based on the median and the median covariation matrix. All the proposed methods are available in the R package RGMM accessible on CRAN.