WSDM22-MCFair

WSDM 2022 paper. Marginal-Certainty-aware Fair Ranking Algorithm. The paper could be found at https://arxiv.org/pdf/2212.09031.pdf

Create the env.

conda env create -f environment.yml

then activate the env

conda activate MCFair

Specify the data directory

We specify data directory in LTRlocal_dataset_info.txt.

Then you can run the default setting

python main.py

and specify the arguments accordingly.

How to launch experiments of different settings?

we provide scripts to run multiple experiments. Set MQ2008 as an example, firstly, we run the following to generate the json settings.

python scripts/datascriptsGradFairLTR/generatingSettingMQ2008.py

Then, you can submit the whole MQ2008 experiments

slurm_python --CODE_PATH=.  --Cmd_file=main.py --JSON_PATH=localOutput/MQ2008  --jobs_limit=10  --secs_each_sub=5 --json2args  --plain_scripts  --only_unfinished

Here slurm_python is a tool to submit multiple jobs and it will be automatically installed when you create the conda env. You can change jobs_limit to set how many jobs to run at the same time. If you have slurm installed in you server, you can remove --plain_scripts to let slurm schedule the jobs.

if you use MCFair in your research, please use the following BibTex entry.

@inproceedings{10.1145/3539597.3570474, author = {Yang, Tao and Xu, Zhichao and Wang, Zhenduo and Tran, Anh and Ai, Qingyao}, title = {Marginal-Certainty-Aware Fair Ranking Algorithm}, year = {2023}, isbn = {9781450394079}, publisher = {Association for Computing Machinery}, address = {New York, NY, USA}, url = {https://doi.org/10.1145/3539597.3570474}, doi = {10.1145/3539597.3570474}, booktitle = {Proceedings of the Sixteenth ACM International Conference on Web Search and Data Mining}, pages = {24–32}, numpages = {9}, series = {WSDM '23} }