A new non-archimedean metric on persistent homology


Creative Commons License

Güzel İ., Kaygun A.

COMPUTATIONAL STATISTICS, vol.37, no.4, pp.1963-1983, 2022 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 37 Issue: 4
  • Publication Date: 2022
  • Doi Number: 10.1007/s00180-021-01187-z
  • Journal Name: COMPUTATIONAL STATISTICS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, ABI/INFORM, zbMATH, Civil Engineering Abstracts
  • Page Numbers: pp.1963-1983
  • Middle East Technical University Affiliated: No

Abstract

In this article, we define a new non-archimedean metric structure, called cophenetic metric, on persistent homology classes of all degrees. We then show that zeroth persistent homology together with the cophenetic metric and hierarchical clustering algorithms with a number of different metrics do deliver statistically verifiable commensurate topological information based on experimental results we obtained on different datasets. We also observe that the resulting clusters coming from cophenetic distance do shine in terms of different evaluation measures such as silhouette score and the Rand index. Moreover, since the cophenetic metric is defined for all homology degrees, one can now display the inter-relations of persistent homology classes in all degrees via rooted trees.