ClustEm4Ano: Clustering Text Embeddings of Nominal Textual Attributes for Microdata Anonymization

TL;DR

ClustEm4Ano is a novel anonymization pipeline that automatically generates semantic value generalization hierarchies for nominal textual data using embeddings and clustering, improving data privacy and utility.

Contribution

It introduces an automated method for creating value generalization hierarchies via clustering of text embeddings, enhancing anonymization of nominal data.

Findings

VGHs generated outperform manual hierarchies in anonymization tasks.

The approach improves data utility for small k-anonymity levels.

Experimental validation on the Adult dataset confirms effectiveness.

Abstract

This work introduces ClustEm4Ano, an anonymization pipeline that can be used for generalization and suppression-based anonymization of nominal textual tabular data. It automatically generates value generalization hierarchies (VGHs) that, in turn, can be used to generalize attributes in quasi-identifiers. The pipeline leverages embeddings to generate semantically close value generalizations through iterative clustering. We applied KMeans and Hierarchical Agglomerative Clustering on $13$ different predefined text embeddings (both open and closed-source (via APIs)). Our approach is experimentally tested on a well-known benchmark dataset for anonymization: The UCI Machine Learning Repository's Adult dataset. ClustEm4Ano supports anonymization procedures by offering more possibilities compared to using arbitrarily chosen VGHs. Experiments demonstrate that these VGHs can outperform manually constructed ones in terms of downstream efficacy (especially for small $k$-anonymity ($2 \leq k \leq 30$)) and therefore can foster the quality of anonymized datasets. Our implementation is made public.