Fair Causal Feature Selection
This addresses fair classification for decision-making tasks, offering a causal approach to improve fairness, though it appears incremental as it builds on existing feature selection methods.
The paper tackled the problem of fair feature selection by proposing FairCFS, which uses localized causal graphs to block sensitive information transmission, achieving comparable accuracy to eight state-of-the-art algorithms and superior fairness on seven real-world datasets.
Fair feature selection for classification decision tasks has recently garnered significant attention from researchers. However, existing fair feature selection algorithms fall short of providing a full explanation of the causal relationship between features and sensitive attributes, potentially impacting the accuracy of fair feature identification. To address this issue, we propose a Fair Causal Feature Selection algorithm, called FairCFS. Specifically, FairCFS constructs a localized causal graph that identifies the Markov blankets of class and sensitive variables, to block the transmission of sensitive information for selecting fair causal features. Extensive experiments on seven public real-world datasets validate that FairCFS has comparable accuracy compared to eight state-of-the-art feature selection algorithms, while presenting more superior fairness.