Nothing
discoCVI: Implementation of the DISCO Metric for Internal Clustering Evaluation
Implementation of the DISCO (Density-based Internal Score for Clusterings with nOise) metric, a cluster validity index for evaluating density-based clustering results without ground truth labels. DISCO is the first index to explicitly assess the quality of noise point assignments in addition to cluster quality. It uses density-connectivity distance derived from a minimum spanning tree of the mutual-reachability graph, providing interpretable, bounded scores in [-1, 1]. Higher scores indicate better clustering. Based on Beer et al. (2025) <doi:10.48550/arXiv.2503.00127>.
Getting started
Browse package contents
Package details |
|
|---|---|
| Author | Amin Entezari [aut, cre] (ORCID: <https://orcid.org/0009-0005-4733-6032>), Davide Chicco [ctb] (ORCID: <https://orcid.org/0000-0001-9655-7142>, Supervision and guidance), Anna Beer [aut] (Original Python implementation (arXiv:2503.00127)), Lena Krieger [aut] (Original Python implementation (arXiv:2503.00127)), Pascal Weber [aut] (Original Python implementation (arXiv:2503.00127)) |
| Maintainer | Amin Entezari <amin_entezari@outlook.com> |
| License | MIT + file LICENSE |
| Version | 0.2.0 |
| URL | https://github.com/aminentezari/discoCVI |
| Package repository | View on CRAN |
| Installation |
Install the latest version of this package by entering the following in R:
|
Try the discoCVI package in your browser
Any scripts or data that you put into this service are public.
