Backbone extraction through statistical edge filtering: A comparative study
2025

Comparative Study of Backbone Extraction Methods

Sample size: 27 publication 10 minutes Evidence: moderate

Author Information

Author(s): Yassin Ali, Cherifi Hocine, Seba Hamida, Togni Olivier

Primary Institution: Université de Bourgogne, Franche-Comté, Dijon, France

Hypothesis

This study aims to systematically compare seven statistical backbone edge filtering methods across diverse networks.

Conclusion

The study provides valuable insights for selecting appropriate backbone extraction methods based on specific properties.

Supporting Evidence

  • The study systematically compares seven influential statistical hypothesis-testing backbone edge filtering methods.
  • Results suggest that the Disparity Filter and LANS filters favor high-weighted edges.
  • The ECM filter assigns lower significance to edges with high degrees.
  • Findings reveal that the Polya Urn filter captures the original weight distribution effectively.
  • Statistical methods maintain network connectivity and weight entropy better than others.

Takeaway

This study looks at different ways to simplify complex networks while keeping important connections, helping researchers choose the best method for their needs.

Methodology

The study compares seven statistical backbone extraction methods using a dataset of 27 real-world networks, analyzing their performance based on various network properties.

Potential Biases

Potential biases may arise from the selection of networks and the specific characteristics of the filtering methods.

Limitations

The study focuses solely on statistical filtering techniques and does not include structural methods.

Statistical Information

P-Value

p<0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1371/journal.pone.0316141

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication