Data ultrametricity and clusterability
Machine Learning
2020-01-08 v1 Machine Learning
Abstract
The increasing needs of clustering massive datasets and the high cost of running clustering algorithms poses difficult problems for users. In this context it is important to determine if a data set is clusterable, that is, it may be partitioned efficiently into well-differentiated groups containing similar objects. We approach data clusterability from an ultrametric-based perspective. A novel approach to determine the ultrametricity of a dataset is proposed via a special type of matrix product, which allows us to evaluate the clusterability of the dataset. Furthermore, we show that by applying our technique to a dissimilarity space will generate the sub-dominant ultrametric of the dissimilarity.
Cite
@article{arxiv.1908.10833,
title = {Data ultrametricity and clusterability},
author = {Dan Simovici and Kaixun Hua},
journal= {arXiv preprint arXiv:1908.10833},
year = {2020}
}