D-BIAS: A Causality-Based Human-in-the-Loop System for Tackling Algorithmic Bias
Open accessThe DOI matches its Crossref record; 22 of the 22 references it cites were found in a registry or an index.
Abstract
With the rise of AI, algorithms have become better at learning underlying patterns from the training data including ingrained social biases based on gender, race, etc. Deployment of such algorithms to domains such as hiring, healthcare, law enforcement, etc. has raised serious concerns about fairness, accountability, trust and interpretability in machine learning algorithms. To alleviate this problem, we propose D-BIAS, a visual interactive tool that embodies human-in-the-loop AI approach for auditing and mitigating social biases from tabular datasets. It uses a graphical causal model to represent causal relationships among different features in the dataset and as a medium to inject domain knowledge. A user can detect the presence of bias against a group, say females, or a subgroup, say black females, by identifying unfair causal relationships in the causal network and using an array of fairness metrics. Thereafter, the user can mitigate bias by refining the causal model and acting on the unfair causal edges. For each interaction, say weakening/deleting a biased causal edge, the system uses a novel method to simulate a new (debiased) dataset based on the current causal model while ensuring a minimal change from the original dataset. Users can visually assess the impact of their interactions on different fairness metrics, utility metrics, data distortion, and the underlying data distribution. Once satisfied, they can download the debiased dataset and use it for any downstream application for fairer predictions. We evaluate D-BIAS by conducting experiments on 3 datasets and also a formal user study. We found that D-BIAS helps reduce bias significantly compared to the baseline debiasing approach across different fairness metrics while incurring little data distortion and a small loss in utility. Moreover, our human-in-the-loop based approach significantly outperforms an automated approach on trust, interpretability and accountability.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Development of NASA-TLX (Task Load Index): Results of Empirical and Theoretical Research
(1988) · Advances in Psychology · doi:10.1016/s0166-4115(08)62386-9 · Cited by 15,003
SUS: A 'Quick and Dirty' Usability Scale
(1996) · Usability Evaluation In Industry · doi:10.1201/9781498710411-35 · Cited by 8,989
Developing and validating trust measures for e-commerce: an integrative typology
McKnight, D. H., Choudhury, V., & Kacmar, C. (2002) · Information Systems Research · doi:10.1287/isre.13.3.334.81 · Cited by 5,550 · cited 2×
“For measuring trust, we referred to McKnight’s framework on Trust [38,39] and other studies [17,22] to come up with 3 questions for our specific case.”
MethodCausation, Prediction, and Search
Spirtes, P., Glymour, C., & Scheines, R. (2001) · doi:10.7551/mitpress/1754.001.0001 · Cited by 4,521
Causal inference in statistics: An overview
Pearl, J. (2009) · Statistics Surveys · doi:10.1214/09-ss057 · Cited by 2,336 · cited 2×
“A direct path is certainly unfair but an indirect path may be fair or unfair (as in the case of Berkeley Admission dataset [35]).”
BackgroundBellamy, R. K. E., Dey, K., Hind, M., Hoffman, S. C., Houde, S., Kannan, K., Lohia, P., Martino, J., Mehta, S., Mojsilovic, A., et al. (2018) · arXiv · doi:10.48550/arxiv.1810.01943 · Cited by 1,581 · cited 3×
“This led us to IBM’s AI Fairness 360 [5] toolkit whose visual interface can be publicly accessed online1.”
BackgroundMethodData preprocessing techniques for classification without discrimination
Kamiran, F., & Calders, T. (2012) · Knowledge and Information Systems · doi:10.1007/s10115-011-0463-8 · Cited by 1,305 · cited 1×
“This can be achieved by either modifying the output label [31,32] or by modifying the input attributes [10,58].”
Initial trust formation in new organizational relationships
McKnight, D. H., Cummings, L. L., & Chervany, N. L. (1998) · The Academy of Management Review · doi:10.2307/259290 · Cited by 1,296
Dietvorst, B. J., Simmons, J. P., & Massey, C. (2018) · Management Science · doi:10.1287/mnsc.2016.2643 · Cited by 1,230
Kusner, M. J., Loftus, J. R., Russell, C., & Silva, R. (2017) · arXiv · doi:10.48550/arxiv.1703.06856 · Cited by 995 · cited 1×
“Numerous fairness metrics have been proposed which capture different facets of fairness, such as group fairness, individual fairness, counterfactual fairness, etc. [4,5,18,34,40].”
BackgroundInherent trade-offs in the fair determination of risk scores
Kleinberg, J., Mullainathan, S., & Raghavan, M. (2017) · Schloss Dagstuhl – Leibniz-Zentrum für Informatik · doi:10.4230/lipics.itcs.2017.43 · Cited by 943 · cited 1×
“, it is impossible to optimize different metrics simultaneously [27].”
BackgroundInherent Trade-Offs in the Fair Determination of Risk Scores
Kleinberg, J., Mullainathan, S., & Raghavan, M. (2016) · arXiv · doi:10.48550/arxiv.1609.05807 · Cited by 941
AI Fairness 360: An extensible toolkit for detecting and mitigating algorithmic bias
Bellamy, R. K. E., Dey, K., Hind, M., Hoffman, S. C., Houde, S., Kannan, K., Lohia, P., Martino, J., Mehta, S., Mojsilović, A., et al. (2019) · IBM Journal of Research and Development · doi:10.1147/jrd.2019.2942287 · Cited by 884
Bickel, P. J., Hammel, E. A., & O'Connell, J. W. (1975) · Science · doi:10.1126/science.187.4175.398 · Cited by 659 · cited 2×
“After correcting for this factor, it was found that the admission process had a small but statistically significant bias in favor of women [6].”
BackgroundMethodClassifying without discriminating
Kamiran, F., & Calders, T. (2009) · 2009 2nd International Conference on Computer, Control and Communication · doi:10.1109/ic4.2009.4909197 · Cited by 409 · cited 2×
“In the pre-processing stage, the dataset is modified such that the bias with respect to an attribute or set of attributes is reduced or eliminated [10, 22, 25, 36, 48].”
BackgroundWhy fairness cannot be automated: Bridging the gap between EU non-discrimination law and AI
Wachter, S., Mittelstadt, B., & Russell, C. (2021) · Computer Law & Security Review · doi:10.1016/j.clsr.2021.105567 · Cited by 336
A Methodology for Direct and Indirect Discrimination Prevention in Data Mining
Hajian, S., & Domingo-Ferrer, J. (2013) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2012.72 · Cited by 312
The effects of example-based explanations in a machine learning interface
Cai, C. J., Jongejan, J., & Holbrook, J. (2019) · Proceedings of the 24th International Conference on Intelligent User Interfaces · doi:10.1145/3301275.3302289 · Cited by 220 · cited 1×
“To capture accountability, we asked two questions based on previous studies [9, 19].”
MethodFairSight: Visual Analytics for Fairness in Decision Making
Ahn, Y., & Lin, Y.-R. (2019) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2019.2934262 · Cited by 126
Drozdal, J., Weisz, J., Wang, D., Dass, G., Yao, B., Zhao, C., Muller, M., Ju, L., & Su, H. (2020) · Proceedings of the 25th International Conference on Intelligent User Interfaces · doi:10.1145/3377325.3377501 · Cited by 119
Silva: Interactively Assessing Machine Learning Fairness Using Causality
Yan, J. N., Gu, Z., Lin, H., & Rzeszotarski, J. M. (2020) · Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3313831.3376447 · Cited by 54
Anti-discrimination learning: a causal modeling-based framework
Zhang, L., & Wu, X. (2017) · International Journal of Data Science and Analytics · doi:10.1007/s41060-017-0058-x · Cited by 48 · cited 1×
“We have made use of causal models since discrimination is inherently causal, and causal models can also be easy to interpret [28,56,59].”
Method
Cited by
Bias and ethics of AI systems applied in auditing - A systematic review
Murikah, W., Nthenge, J. K., & Musyoka, F. M. (2024) · Scientific African · doi:10.1016/j.sciaf.2024.e02281 · Cited by 148
Guiding principles for the responsible development of artificial intelligence tools for healthcare
Badal, K., Lee, C. M., & Esserman, L. J. (2023) · Communications Medicine · doi:10.1038/s43856-023-00279-9 · Cited by 77
Bias Mitigation in Primary Health Care Artificial Intelligence Models: Scoping Review
Sasseville, M., Ouellet, S., Rhéaume, C., Sahlia, M., Couture, V., Després, P., Paquette, J.-S., Darmon, D., Bergeron, F., & Gagnon, M.-P. (2025) · Journal of Medical Internet Research · doi:10.2196/60269 · Cited by 38
STILE: Exploring and Debugging Social Biases in Pre-trained Text Representations
Kabir, S., Li, L., & Zhang, T. (2024) · Proceedings of the CHI Conference on Human Factors in Computing Systems · doi:10.1145/3613904.3642111 · Cited by 8
Metadata
- Authors
- Ghai, B., & Mueller, K.
- Venue
- IEEE Transactions on Visualization and Computer Graphics
- Published
- 2022
- Volume
- 29
- Issue
- 1
- Pages
- 1-10
- ISSN
- 1077-2626 (ISSN-L)
- 1941-0506
- 2160-9306
- OpenAlex
- W4297094586
- Cited by
- 52
- References
- 22
- FWCI
- 13.08
- Type
- journal article
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.