Deciphering Network Community Structure by Surprise Network Community Detection
2011

Understanding Community Structure in Complex Networks

publication Evidence: high

Author Information

Author(s): Aldecoa Rodrigo, Marín Ignacio

Primary Institution: Instituto de Biomedicina de Valencia, Consejo Superior de Investigaciones Científicas, Valencia, Spain

Hypothesis

Can the global parameter Surprise (S) provide a better characterization of community structure in complex networks compared to Newman and Girvan's modularity (Q)?

Conclusion

Maximizing the Surprise parameter provides a more effective characterization of community structures in complex networks than traditional methods.

Supporting Evidence

  • Maximizing Surprise often leads to optimal characterizations of existing communities.
  • Surprise outperforms modularity in detecting community structures in various benchmarks.
  • Real networks analyzed showed that Surprise can accurately reflect known community structures.

Takeaway

This study shows that using a new method called Surprise helps us find groups in complex networks better than older methods.

Methodology

The study used both standard and novel benchmarks to test the performance of the Surprise parameter against traditional methods.

Potential Biases

There may be psychological biases affecting the evaluation of community structures in networks.

Limitations

The algorithms used may not always detect the maximum possible Surprise values, leading to potential underestimation of its effectiveness.

Digital Object Identifier (DOI)

10.1371/journal.pone.0024195

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication