A Conceptual Framework for Assessing Anonymization-Utility Trade-Offs Based on Principal Component Analysis

TL;DR

This paper introduces a PCA-based anonymization method for databases that balances data privacy with utility, using novel metrics to evaluate the utility of anonymized data.

Contribution

It presents a new framework integrating PCA with utility metrics to optimize data anonymization while maintaining query usefulness.

Findings

Effective utility measurement methods proposed

Framework successfully balances privacy and data utility

Applicable to various database types and query responses

Abstract

An anonymization technique for databases is proposed that employs Principal Component Analysis. The technique aims at releasing the least possible amount of information, while preserving the utility of the data released in response to queries. The general scheme is described, and alternative metrics are proposed to assess utility, based respectively on matrix norms; correlation coefficients; divergence measures, and quality indices of database images. This approach allows to properly measure the utility of output data and incorporate that measure in the anonymization method.