Saved in:
Bibliographic Details
Main Authors: Gagolewski, Marek, Cena, Anna, Bartoszuk, Maciej, Brzozowski, Łukasz
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2303.05679
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908592997662720
author Gagolewski, Marek
Cena, Anna
Bartoszuk, Maciej
Brzozowski, Łukasz
author_facet Gagolewski, Marek
Cena, Anna
Bartoszuk, Maciej
Brzozowski, Łukasz
contents Minimum spanning trees (MSTs) provide a convenient representation of datasets in numerous pattern recognition activities. Moreover, they are relatively fast to compute. In this paper, we quantify the extent to which they are meaningful in low-dimensional partitional data clustering tasks. By identifying the upper bounds for the agreement between the best (oracle) algorithm and the expert labels from a large battery of benchmark data, we discover that MST methods can be very competitive. Next, we review, study, extend, and generalise a few existing, state-of-the-art MST-based partitioning schemes. This leads to some new noteworthy approaches. Overall, the Genie and the information-theoretic methods often outperform the non-MST algorithms such as K-means, Gaussian mixtures, spectral clustering, Birch, density-based, and classical hierarchical agglomerative procedures. Nevertheless, we identify that there is still some room for improvement, and thus the development of novel algorithms is encouraged.
format Preprint
id arxiv_https___arxiv_org_abs_2303_05679
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Clustering with minimum spanning trees: How good can it be?
Gagolewski, Marek
Cena, Anna
Bartoszuk, Maciej
Brzozowski, Łukasz
Machine Learning
Minimum spanning trees (MSTs) provide a convenient representation of datasets in numerous pattern recognition activities. Moreover, they are relatively fast to compute. In this paper, we quantify the extent to which they are meaningful in low-dimensional partitional data clustering tasks. By identifying the upper bounds for the agreement between the best (oracle) algorithm and the expert labels from a large battery of benchmark data, we discover that MST methods can be very competitive. Next, we review, study, extend, and generalise a few existing, state-of-the-art MST-based partitioning schemes. This leads to some new noteworthy approaches. Overall, the Genie and the information-theoretic methods often outperform the non-MST algorithms such as K-means, Gaussian mixtures, spectral clustering, Birch, density-based, and classical hierarchical agglomerative procedures. Nevertheless, we identify that there is still some room for improvement, and thus the development of novel algorithms is encouraged.
title Clustering with minimum spanning trees: How good can it be?
topic Machine Learning
url https://arxiv.org/abs/2303.05679