This repository contains the code to conduct the analysis of the article: Rajeh, S., Savonnet, M., Leclercq, E. et al. Characterizing the interactions between classical and community-aware centrality measures in complex networks. Sci Rep 11, 10088 (2021). https://doi.org/10.1038/s41598-021-89549-x
Comments and questions are welcome, contact: stephany.rajeh(at)u-bourgogne.fr
The sources of datasets used in the study are available within the article.
- Calculates the classical and community-aware centrality for a given network.
- Computes the correlation for all possible combinations between classical and community-aware centrality measures are then represent them in a heatmap.
Note: Classical measures are already written with networkx
in Python and centiserve
in R while community-aware centrality measures are written for the study of this paper.
- Extracts macroscopic features for a given network
- Extracts mesoscopic features for a given network
Note: Code for macroscopic features is already written with networkx
in Python while code for mesoscopic features are written for the study of this paper.
- Code for performing linear regression using ordinary least squares (in Python)
- Code for performing linear regression using weighted least squares (in R)
Note: Linear regression concering the Degree Distribution Exponent is calculated excluding the Football network, in a separate folder.