p-value Adjustment for Monotonous, Unbiased, and Fast Clustering Comparison

Klede K, Altstidl TR, Zanca D, Eskofier B (2023)

Publication Language: English

Publication Type: Conference contribution, Original article

Publication year: 2023

Publisher: Curran Associates, Inc.

Series: Advances in Neural Information Processing Systems

Book Volume: 36

Pages Range: 27113–27128

Conference Proceedings Title: Advances in Neural Information Processing Systems 36

Event location: New Orleans US

URI: https://proceedings.neurips.cc/paper_files/paper/2023/file/563d94819f68cb73d6a382809e587b54-Paper-Conference.pdf

Open Access Link: https://proceedings.neurips.cc/paper_files/paper/2023/file/563d94819f68cb73d6a382809e587b54-Paper-Conference.pdf

Abstract

Popular metrics for clustering comparison, like the Adjusted Rand Index and the Adjusted Mutual Information, are type II biased. The Standardized Mutual Information removes this bias but suffers from counterintuitive non-monotonicity and poor computational efficiency. We introduce the p-value adjusted Rand Index (PMI2), the first cluster comparison method that is type II unbiased and provably monotonous. The PMI2 has fast approximations that outperform the Standardized Mutual information. We demonstrate its unbiased clustering selection, approximation quality, and runtime efficiency on synthetic benchmarks. In experiments on image and social network datasets, we show how the PMI2 can help practitioners choose better clustering and community detection algorithms.

Authors with CRIS profile

Kai Klede Machine Learning and Data Analytics Lab Thomas Robert Altstidl Machine Learning and Data Analytics Lab Dario Zanca Machine Learning and Data Analytics Lab Björn Eskofier Machine Learning and Data Analytics Lab

How to cite

APA:

Klede, K., Altstidl, T.R., Zanca, D., & Eskofier, B. (2023). p-value Adjustment for Monotonous, Unbiased, and Fast Clustering Comparison. In A. Oh and T. Neumann and A. Globerson and K. Saenko and M. Hardt and S. Levine (Eds.), Advances in Neural Information Processing Systems 36 (pp. 27113–27128). New Orleans, US: Curran Associates, Inc..

MLA:

Klede, Kai, et al. "p-value Adjustment for Monotonous, Unbiased, and Fast Clustering Comparison." Proceedings of the Thirty-seventh Annual Conference on Neural Information Processing Systems, New Orleans Ed. A. Oh and T. Neumann and A. Globerson and K. Saenko and M. Hardt and S. Levine, Curran Associates, Inc., 2023. 27113–27128.

BibTeX: Download