HTF: Homogeneous Tree Framework for Differentially-Private Release of Location Data

TL;DR

This paper introduces HTF, a novel data structure that improves the accuracy of differentially-private location histograms by partitioning data into homogeneous density regions, validated through extensive real-world experiments.

Contribution

The paper proposes a homogeneous density-based tree framework (HTF) for better differential privacy histogram accuracy, addressing limitations of existing data-dependent methods.

Findings

HTF outperforms existing DP histogram methods in accuracy.

Density homogeneity is key to improving DP histogram utility.

Extensive experiments confirm the effectiveness of HTF on real-world data.

Abstract

Mobile apps that use location data are pervasive, spanning domains such as transportation, urban planning and healthcare. Important use cases for location data rely on statistical queries, e.g., identifying hotspots where users work and travel. Such queries can be answered efficiently by building histograms. However, precise histograms can expose sensitive details about individual users. Differential privacy (DP) is a mature and widely-adopted protection model, but most approaches for DP-compliant histograms work in a data-independent fashion, leading to poor accuracy. The few proposed data-dependent techniques attempt to adjust histogram partitions based on dataset characteristics, but they do not perform well due to the addition of noise required to achieve DP. We identify density homogeneity as a main factor driving the accuracy of DP-compliant histograms, and we build a data structure that splits the space such that data density is homogeneous within each resulting partition. We show through extensive experiments on large-scale real-world data that the proposed approach achieves superior accuracy compared to existing approaches.