English

Quantile-based clustering

Methodology 2019-11-12 v2

Abstract

A new cluster analysis method, KK-quantiles clustering, is introduced. KK-quantiles clustering can be computed by a simple greedy algorithm in the style of the classical Lloyd's algorithm for KK-means. It can be applied to large and high-dimensional datasets. It allows for within-cluster skewness and internal variable scaling based on within-cluster variation. Different versions allow for different levels of parsimony and computational efficiency. Although KK-quantiles clustering is conceived as nonparametric, it can be connected to a fixed partition model of generalized asymmetric Laplace-distributions. The consistency of KK-quantiles clustering is proved, and it is shown that KK-quantiles clusters correspond to well separated mixture components in a nonparametric mixture. In a simulation, KK-quantiles clustering is compared with a number of popular clustering methods with good results. A high-dimensional microarray dataset is clustered by KK-quantiles.

Keywords

Cite

@article{arxiv.1806.10403,
  title  = {Quantile-based clustering},
  author = {Christian Hennig and Cinzia Viroli and Laura Anderlucci},
  journal= {arXiv preprint arXiv:1806.10403},
  year   = {2019}
}