Quantifying the Re-identification Risk in Published Process Models

Abstract Event logs are the basis of process mining operations such as discovery, conformance checking, and optimization. Sensitive information may be obtained by adversaries when re-identifying individuals that relate to traces an event log. This re-identification risk is dependent on assumed background attacker. Multiple techniques have been proposed quantify risks for published logs. However, in many scenarios there no need release full log, a discovered model annotated with frequencies suffices. raises question how models. We propose method based generating sample this trees frequencies. The was applied several real-life Inductive Miner. Our results show can still significant publishing tree; however, often lower than releasing original