Bias analysis and mitigation in data-driven tools using provenance

Yuval Moskovitch, Jinyang Li, H. V. Jagadish

Abstract: Fairness and bias mitigation in data-driven systems has been extensively studied in recent years. In this paper, we suggest a novel approach towards fairness analysis and bias mitigation utilizing the notion of provenance, which was shown to be useful for similar tasks in the context of data and process analyses. We illustrate the idea using a simple use-case demonstrating a scenario of mitigating bias caused by inadequate minority group representation. We conclude with an outline of opportunities and challenges in developing provenance-based solutions for bias analysis and mitigation in data-driven systems.