Fairness in a real social network

This is a project for measuring fairness in ranking algorithms results. It was developed for the course "D6 - Online Social Networks and Media" of my MSc degree under the supervision of E. Pitoura and P. Tsaparas. A full description of the process followed and results can be found in REPORT.pdf file.

Notes

• requirements.txt file can be used to re-create anaconda environment with the necessary dependencies.

Contents

• src/fairness_funcs.py: Fairness-related metrics calculations such as GINI coefficient or homophily
• src/helper_funcs.py: Graph preprocessing and I/O functions
• gplus_analysis.ipynb: Fairness analysis of Google+ ego-networks

References

1. Tóth, G., Wachs, J., Di Clemente, R. et al. Inequality is rising where social network segregation interacts with urban topology. Nat Commun 12, 1143 (2021). https://doi.org/10.1038/s41467-021-21465-0
2. J. McAuley and J. Leskovec. Learning to Discover Social Circles in Ego Networks. NIPS, 2012. https://snap.stanford.edu/data/ego-Gplus.html