Beyond the Trade-off Curve: Multivariate and Advanced Risk-Utility Maps for Evaluating Anonymized and Synthetic Data

TL;DR

This paper compares six multivariate visualization methods to evaluate anonymized data, enabling more informed decisions by revealing complex risk-utility relationships.

Contribution

It introduces new PCA-based visualization techniques and systematically assesses their effectiveness for multivariate risk-utility analysis.

Findings

Blockwise PCA enhances composite scatterplots.

Joint PCA reveals measure interrelationships.

Multivariate visualization improves method selection.

Abstract

Anonymizing microdata requires balancing the reduction of disclosure risk with the preservation of data utility. Traditional evaluations often rely on single measures or two-dimensional risk-utility (R-U) maps, but real-world assessments involve multiple, often correlated, indicators of both risk and utility. Pairwise comparisons of these measures can be inefficient and incomplete. We therefore systematically compare six visualization approaches for simultaneous evaluation of multiple risk and utility measures: heatmaps, dot plots, composite scatterplots, parallel coordinate plots, radial profile charts, and PCA-based biplots. We introduce blockwise PCA for composite scatterplots and joint PCA for biplots that simultaneously reveal method performance and measure interrelationships. Through systematic identification of Pareto-optimal methods in all approaches, we demonstrate how multivariate visualization supports a more informed selection of anonymization methods.