Diverse Community Data for Benchmarking Data Privacy Algorithms
This work addresses bias and privacy challenges in data deidentification for researchers and practitioners, but it is incremental as it builds on existing benchmarking efforts.
The paper tackles the problem of bias and privacy issues in tabular data deidentification algorithms by summarizing contributions from the Collaborative Research Cycle, including theoretical work, public benchmark data, evaluation tools, and an archive of over 450 deidentified data samples, with initial results showing the value of these tools for investigations.
The Collaborative Research Cycle (CRC) is a National Institute of Standards and Technology (NIST) benchmarking program intended to strengthen understanding of tabular data deidentification technologies. Deidentification algorithms are vulnerable to the same bias and privacy issues that impact other data analytics and machine learning applications, and can even amplify those issues by contaminating downstream applications. This paper summarizes four CRC contributions: theoretical work on the relationship between diverse populations and challenges for equitable deidentification; public benchmark data focused on diverse populations and challenging features; a comprehensive open source suite of evaluation metrology for deidentified datasets; and an archive of more than 450 deidentified data samples from a broad range of techniques. The initial set of evaluation results demonstrate the value of these tools for investigations in this field.