Fair-Multi-Criteria-Candidate-Set-Selection

Corresponding code, and experiments for paper "Fair & Share: Fast and Fair Multi-Criteria Selections".

Fair&Share and F&S-Fagin's code:

Please see src folder for FairShare.py python script for Fair&Share.

F&S-Fagin's code is in the fair_fagins.py script in the baselines folder.

Comparative baselines:

All baselines used are in the baselines folder.

• Fair. Please see baseline_zehlikeetal.py.
• DivTopK. Please see baseline_stoyanovichetal.py.
• GbG. Please see baseline_GBG_Threshold.py.
• Greedy. Please see baseline_greedyfmc.py.

Experiments:

Proportional Representation Task

For experiments with proportional fairness objectives please see the exp_with_proportional folder. There study_delta_proportional.py performs the experiments. All results are then in the exp_with_proportionalfolder.

Equal Representation Task

For experiments with equal fairness objectives please see the exp_with_equal folder. There study_delta_equal.py performs the experiments. All results are then in the exp_with_equal folder.

Rooney Rule Task

For experiments with the rooney rule please see the exp_with_rooney folder. There run_rooney.py performs the experiments comparing all methods. Results are then in the rooney_taskadult.csv file.

Helper Functions:

Functions to calculate fairness metrics are in the metrics directory.