Abstract—Many companies, non-profit organizations and governmental bodies collect personal information during service interactions. However, releasing sensitive personal data may impose huge privacy risks. First, an increasing amount of sensitive personal information becomes publicly available online after user consent. Moreover, data breaches may result in huge data dumps that can contain personal records of millions of individuals. Hence, malicious entities are able to scrape, collect and combine personal data from multiple sources in order to compile detailed profiles of many individuals. This paper demonstrates the impact of publicly available data during de-anonymization by means of a concrete case study. Journalists are often reluctant or even prohibited to release the identity of suspects or victims in criminal cases. They do, however, often release initials and background (such as their age and residential location). Through a large scale study of over 132.000 news articles, this paper demonstrates that currently applied privacy measures are often insufficient and straightforward re-identification strategies can de-anonymize individuals.

The poster was accepted at the 7th IEEE European Symposium on Security and Privacy (Euro S&P 2022) and presented in the poster session. Original poster: https://ieeeeurosp.github.io/2022/posters/