Skip to content

Repository of the paper "Community detection in bipartite signed networks is highly dependent on parameter choice"

Notifications You must be signed in to change notification settings

elenacandellone/signed-bipartite-nets

Repository files navigation

Community detection on bipartite signed networks

Repository for the paper "Community detection in bipartite signed networks is highly dependent on parameter choice". https://arxiv.org/abs/2405.08203


Paper abstract

Decision-making processes often involve voting. Human interactions with exogenous entities such as legislations or products can be effectively modeled as two-mode (bipartite) signed networks---where people can either vote positively, negatively, or abstain from voting on the entities. Detecting communities in such networks could help us understand underlying properties: for example ideological camps or consumer preferences. While community detection is an established practice separately for bipartite and signed networks, it remains largely unexplored in the case of bipartite signed networks. In this paper, we systematically evaluate the efficacy of community detection methods on bipartite signed networks using a synthetic benchmark and real-world datasets. Our findings reveal that when no communities are present in the data, these methods often recover spurious communities. When communities are present, the algorithms exhibit promising performance, although their performance is highly susceptible to parameter choice. This indicates that researchers using community detection methods in the context of bipartite signed networks should not take the communities found at face value: it is essential to assess the robustness of parameter choices or perform domain-specific external validation.

Synthetic Networks Benchmark

Repository installation

Clone this repository with the command

 git clone https://github.com/elenacandellone/signed-bipartite-nets.git

Install the required packages

pip install -r requirements.txt

Scripts

  • 1-us-house-scrape-data.py scrapes data from the website of the US House of Representatives Clerk and saves the votes in the data folder.
  • 2a-us-house-covoting.py creates the co-voting network for the US House of Representatives data.
  • 2b-meneame-covoting.py creates the co-voting network for the Meneame data.
  • 3-synth-nets.ipynb generates the synthetic networks based on the four controlled scenarios and insights from real data.
  • 4-community-detection.py performs the community detection using several methods on both real and synthetic networks.
  • 5a-analysis.ipynb results of the clustering evaluation for community-spinglass and SPONGE.
  • 5b-analysis-sbm.ipynb results of the clustering evaluation for SBM.

Presentations

  • ODISSEI Conference 2023 DOI
  • Young Complexity Researchers Utrecht [link]
  • NetSci 2024 DOI

How to cite

E. Candellone, E. van Kesteren, S.Chelmi, J. Garcia Bernardo. Community detection in bipartite signed networks is highly dependent on parameter choice, 2024.

@misc{candellone2024community,
      title={Community detection in bipartite signed networks is highly dependent on parameter choice}, 
      author={Elena Candellone and Erik-Jan van Kesteren and Sofia Chelmi and Javier Garcia-Bernardo},
      year={2024},
      eprint={2405.08203},
      archivePrefix={arXiv},
      doi={https://doi.org/10.48550/arXiv.2405.08203}
}

Contact

Project by the ODISSEI Social Data Science (SoDa) team.

About

Repository of the paper "Community detection in bipartite signed networks is highly dependent on parameter choice"

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published