Releasing Differentially Private Event Logs Using Generative Models

Frederik Wangelik, Majid Rafiei, Mahsa Pourbafrani, Wil M.P. van der Aalst · 2025

Read Releasing Differentially Private Event Logs Using Generative Models by Frederik Wangelik, Majid Rafiei, Mahsa Pourbafrani, Wil M.P. van der Aalst (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-06418.txt.

Description

Frederik Wangelik, Majid Rafiei, Mahsa Pourbafrani, and Wil van der Aalst proposed ways in 2025 to release private event logs using generative models. Process mining analyses event data that often hold sensitive facts. Past privacy work covered trace variants, the distinct paths through a process, but industry needs and rare variants remain hard.

One method trains an adversarial network to sample from a privatised distribution of variants. The second uses diffusion models that rebuild artificial variants from noise. Both avoid faults of older methods, such as capping variant length or adding fake variants. Tests on real event data beat prior methods on privacy and utility together.

Catalog record

CREATORFrederik Wangelik, Majid Rafiei, Mahsa Pourbafrani, Wil M.P. van der Aalst
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE75 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-06418

Browse the full library · About the Tower of Babel