The source code for the papers:
Nikolaos Fanourakis, Christos Kontousias, Vasilis Efthymiou, Vassilis Christophides and Dimitris Plexousakis. FairER demo: Fairness-Aware and Explainable Entity Resolution. ISWC 2023 [pdf] [poster] [demo video]
Vasilis Efthymiou, Kostas Stefanidis, Evaggelia Pitoura, Vassilis Christophides. FairER: Entity Resolution with Fairness Constraints. CIKM 2021 [pdf] [slides] [poster]
which are parts of the project ResponsibleER: Responsible by Design Entity Resolution, funded by the Hellenic Foundation for Research and Innovation.
A video presentation of FairER demo is located in "demo_video/new_demo_video.mkv" of this repository; also available on Youtube.
Public API and UI (demo with restricted functionality):
Before installing and running the code locally, with its full functionality, you may be interested in checking the public version with limited functionality, hosted at https://isl.ics.forth.gr/fairER/. There, you can also find the API ("Services Manual"), along with executable examples.
cd fairER/installation
conda env create --file demo.yml
conda env create --file demo_bert_int.yml
pip install torch==1.11.0+cu113 torchvision==0.12.0+cu113 torchaudio==0.11.0 --extra-index-url https://download.pytorch.org/whl/cu113
conda activate demo
cd ../matching/OpenEA
pip install -e .conda activate demo
export PYTHONPATH=$PYTHONPATH:/home/<username>/fairER/
python web/run.py- Please click the "Download Datasets" button after the successfull installation of the app.
- Download "2016-10-des_dict" from https://github.com/kosugi11037/bert-int and place it to resources/Datasets
- Download "wiki-news-300d-1M.vec" from https://fasttext.cc/docs/en/english-vectors.html and place it to matching/OpenEA/run