The ANO-NET project develops methods for ensuring the anonymity of individuals in social network data. When data on people is analyzed for scientific purposes, privacy of the individuals involved is of the utmost importance. For conventional tabular data, existing methods from the field of statistical disclosure control can guarantee that individuals in the data cannot be identified based on their characteristics.

In the ANO-NET project we aim to develop similar disclosure control approaches, but then for complex network data. The algorithms developed in the project can be used to measure and control the anonymity of individuals, not based on their characteristics, but based on their connections, so their structural position in the social network they are part of. This is a nontrivial process, as assessing network positions relates to well-known computationally challenging problems in graph theory.

The project is a collaboration between Statistics Netherlands (CBS) and Leiden University's Computational Network Science (CNS) group. The outcomes will be useful for researchers and practitioners that want to share social network data while ensuring the anonymity of individuals in this data. In that sense, it helps ensure that regulations regarding privacy and data protection, such as those in the GDPR, are measurably adhered to. More in particular, the project's results will be reusable by sister project POPNET on population-scale social network analysis.

Within this project, one of the aims is to create a durable research infrastructure for analysis of large-scale social network data on the Dutch population. In that context, statistical disclosure control measures are of extreme relevance. The ANO-NET project started in October 2021 and lasts until 2025.

Share on Facebook Share on X Share on LinkedIn Share by WhatsApp Share by Mastodon