CoDeBetHe

This module allows to perform Community Detection with the Bethe Hessian matrix on sparse unweighted and undirected graphs. The module is composed mainly of two algorithms, one for static community detection, following the results presented in A unified framework for spectral clustering in sparse graphs and Revisiting the Bethe-Hessian: Improved Community Detection in Sparse Heterogeneous Graphs and one for dynamic community detection, following Community detection in sparse time-evolving graphs with a dynamical Bethe-Hessian.

If you use CoDeBetHe, please consider to cite the related articles.

This implementations introduce further tricks to make the algorithms more efficent. In order to use CoDeBetHe, open your terminal and type

julia
pkg> add https://github.com/lorenzodallamico/CoDeBetHe

In the next we will show how to use the main packages for static and dynamic community detection.

• CoDeBetHe
• Dynamic community detection
• Create a synthetic graphs
• Load a synthetic data-set
• Infer the community structure from $\{A^{(t)}\}_{t = 1,\dots,T}$
• Static community detection
• Create a synthetic graphs
• Load a real graphs
• Infer the community structure from $A$
• Useful functions