# A Pragmatic Method for Comparing Clusterings with Overlaps and Outliers

arXiv: https://arxiv.org/abs/2602.14855 · HTML: https://mlsanity.com/paper/2602.14855/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 28 references that are in the map.
- First posted on arXiv 16 Feb 2026.

## Builds on (most cited first, 9 of 9)

- [From Louvain to Leiden: guaranteeing well-connected communities](https://mlsanity.com/paper/1810.08473/) (2018)
- [The Impact of Random Models on Clustering Similarity](https://mlsanity.com/paper/1701.06508/) (2017)
- [Artificial Benchmark for Community Detection (ABCD): Fast Random Graph Model with Community Structure](https://mlsanity.com/paper/2002.00843/) (2020)
- [Element-centric clustering comparison unifies overlaps and hierarchy](https://mlsanity.com/paper/1706.06136/) (2017)
- [F*: An Interpretable Transformation of the F-measure](https://mlsanity.com/paper/2008.00103/) (2020)
- [The Artificial Benchmark for Community Detection with Outliers and Overlapping Communities (ABCD+o2)](https://mlsanity.com/paper/2506.05486/) (2025)
- [Comparing Graph Clusterings: Set partition measures vs. Graph-aware measures](https://mlsanity.com/paper/1806.11494/) (2018)
- [Random Models for Fuzzy Clustering Similarity Measures](https://mlsanity.com/paper/2312.10270/) (2023)
- [Hierarchical Single-Linkage Clustering for Community Detection with Overlaps and Outliers](https://mlsanity.com/paper/2509.02334/) (2025)

## Built on by (most cited first, 1 of 1)

- [Leveraging Non-linear Dimension Reduction and Random Walk Co-occurrence for Node Embedding](https://mlsanity.com/paper/2602.24069/) (2026)

## Abstract (arXiv)

Clustering algorithms are an essential part of the unsupervised data science ecosystem, and extrinsic evaluation of clustering algorithms requires a method for comparing the detected clustering to a ground truth clustering. In a general setting, the detected and ground truth clusterings may have outliers (objects belonging to no cluster), overlapping clusters (objects may belong to more than one cluster), or both, but methods for comparing these clusterings are currently undeveloped. In this note, we define a pragmatic similarity measure for comparing clusterings with overlaps and outliers, show that it has several desirable properties, and experimentally confirm that it is not subject to several common biases afflicting other clustering comparison measures.