Algorithmic Fairness Datasets: the Story so Far

TL;DR

This paper surveys over two hundred datasets used in algorithmic fairness research, providing standardized documentation and analysis to address data documentation gaps and improve dataset understanding for fair machine learning.

Contribution

It offers a comprehensive, standardized documentation of key fairness datasets and analyzes their properties, limitations, and ethical considerations, unifying prior scholarship.

Findings

Identified the three most popular fairness datasets: Adult, COMPAS, and German Credit.

Provided detailed documentation and analysis of hundreds of datasets and their properties.

Highlighted best practices for dataset curation in fairness research.

Abstract

Data-driven algorithms are studied in diverse domains to support critical decisions, directly impacting people's well-being. As a result, a growing community of researchers has been investigating the equity of existing algorithms and proposing novel ones, advancing the understanding of risks and opportunities of automated decision-making for historically disadvantaged populations. Progress in fair Machine Learning hinges on data, which can be appropriately used only if adequately documented. Unfortunately, the algorithmic fairness community suffers from a collective data documentation debt caused by a lack of information on specific resources (opacity) and scatteredness of available information (sparsity). In this work, we target data documentation debt by surveying over two hundred datasets employed in algorithmic fairness research, and producing standardized and searchable documentation for each of them. Moreover we rigorously identify the three most popular fairness datasets, namely Adult, COMPAS and German Credit, for which we compile in-depth documentation. This unifying documentation effort supports multiple contributions. Firstly, we summarize the merits and limitations of Adult, COMPAS and German Credit, adding to and unifying recent scholarship, calling into question their suitability as general-purpose fairness benchmarks. Secondly, we document and summarize hundreds of available alternatives, annotating their domain and supported fairness tasks, along with additional properties of interest for fairness researchers. Finally, we analyze these datasets from the perspective of five important data curation topics: anonymization, consent, inclusivity, sensitive attributes, and transparency. We discuss different approaches and levels of attention to these topics, making them tangible, and distill them into a set of best practices for the curation of novel resources.