FanFAIR
Semi-automatic assessment of datasets fairness
What is FanFAIR
FanFAIR is a rule-based approach based on fuzzy logic able to calculate some fairness metrics over a dataset and combine them into a single score, enabling a semi-automatic evaluation of a dataset in algorithmic fairness research.
Using FanFAIR
FanFAIR is designed to be as automatic as possible. However, two metrics (quality, compliance) require human intervention. Here is an example of analysis performed with FanFAIR:
from fanfair import FanFAIR
FF = FanFAIR(dataset="myfile.csv", target_column="output")
FF.set_compliance( {"data_protection_law": True,
"copyright_law": True,
"medical_law": True,
"non_discrimination_law": False,
"ethics": False})
FF.set_quality(0.9)
FF.produce_report()
The analysis is automatically performed by calling the produce_report method, which generates two main figures: the gauge with the overall fairness score (from 0% to 100%), and the plots of the linguistic variables of the fuzzy model, which provide a summary of the metrics for the dataset's fairenss features.
Citing FanFAIR
If you find FanFAIR useful for your research, please cite our project as follows:
Gallese C., Scantamburlo T., Manzoni L., Nobile M.S.: Investigating Semi-Automatic Assessment of Data Sets Fairness by Means of Fuzzy Logic, Proceedings of the 20th IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (IEEE CIBCB 2023), 2023
If you need additional information, or want to see additional metrics implemented in FanFAIR, please feel free to contact Dr. Chiara Gallese (chiara.gallese@unito.it).
Acknowledgements
Metadata
Release files for FanFAIR 1.0.2
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| fanfair-1.0.2.tar.gz | 10.0 kB | Without provenance | Details for fanfair-1.0.2.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| FanFAIR-1.0.2-py3-none-any.whl | Python 3 | none | any | Details for FanFAIR-1.0.2-py3-none-any.whl |
Total release size: 20.3 kB
Release files / fanfair-1.0.2.tar.gz
| Download URL | fanfair-1.0.2.tar.gz |
|---|---|
| Size | 10.0 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
cf13fca094a3a8f56dc4d2042134e162141a7124ab2a93d249945b8684d9496a
|
|
BLAKE2b-256 checksum How to use checksums |
8589c47f6d38d170ba007889b6aebc7ae2803a633e5b2bf65c55550fa65b3743
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/5.1.0 CPython/3.11.7
|
Release files / FanFAIR-1.0.2-py3-none-any.whl
| Download URL | FanFAIR-1.0.2-py3-none-any.whl |
|---|---|
| Size | 10.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
e3a45ad8545ea1941c6191c9724be779ffb87b9556ee17c804e93a6d66858cc7
|
|
BLAKE2b-256 checksum How to use checksums |
8e775cf23dafd6e9ec268c6ea8dde90589da50e2c057443a3c5e0ddc55bf08bf
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/5.1.0 CPython/3.11.7
|
Security for FanFAIR 1.0.2
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.