Geometric description of clustering in directed networks

This repository contains the code associated with Geometric description of clustering in directed networks by A. Allard, M. Á. Serrano and M. Boguñá.

Installation

Installing and running this code requires CMake and Python 3.

1. Clone this repository with the option --recurse-submodules

git clone --recurse-submodules https://github.com/directed-geometric-networks/directedS1.git

2. Copy the directedS1_data subdirectory out of the project's root directory.

cp -R directed-geometric-networks/directed-geometric-networks_data .

3. Compile the executables

cd directed-geometric-networks/analyze_networks; cmake .; make; cd ../..
cd directed-geometric-networks/infer_parameters; cmake .; make; cd ../..

Real-world network data

Real-world network data was downloaded from Netzschleuder. Each dataset was downloaded in the GraphML format (.xml.zst) in the directory directed-geometric-networks_data/real_networks/rough_data/netzschleuder/ and extracted in the proper format by running

python extract_edgelists.py

Generating the results and producing the figures

Details on how the results and the figures were generated are provided in the following notebooks

• plot_figure_real_networks_reciprocity_vs_clustering.ipynb
• plot_figure_reciprocity_real_vs_synthetic.ipynb
• plot_figure_synthetic_networks_global_properties.ipynb
• plot_figure_synthetic_networks_reciprocity_vs_clustering.ipynb
• plot_figure_triangle_spectrum.ipynb
• plot_figure_validation_degree.ipynb
• plot_figure_validation_reciprocity.ipynb

Paper

If you use this code, please consider citing:

Geometric description of clustering in directed networks
A. Allard, M. Á. Serrano and M. Boguñá
arXiv:2302.09055 (2023)