No Community Detection Method to Rule Them All!
Shrabani Ghosh [0000-0002-6084-4964], Erik Saule [0000-0003-1634-9234] · 2025
Read No Community Detection Method to Rule Them All! by Shrabani Ghosh [0000-0002-6084-4964], Erik Saule [0000-0003-1634-9234] (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-11490.txt.
Description
Shrabani Ghosh and Erik Saule ask whether any community detection method is best for later tasks, in a 2025 network science study. Community detection finds groups of tightly linked nodes in large graphs, and its output often feeds later work as extra features.
Past scoring mostly rated methods by built-in goals like modularity, not by what happens downstream. The authors link community traits to task metrics such as precision, recall, and F1 across many methods and two tasks. Thousands of community sets were explored. No single trait explained the scores directly; outcomes came from complex mixes of traits. So no standard method gave the best downstream results in all cases. A method mixing random community making with simple machine learning did better.
Catalog record
| CREATOR | Shrabani Ghosh [0000-0002-6084-4964], Erik Saule [0000-0003-1634-9234] |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 30 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-11490