Measuring Rule-based LTLf Process Specifications: A Probabilistic Data-driven Approach
This work addresses the need for more accurate compliance measurement in process mining for users dealing with event logs, though it is incremental as it builds on existing declarative specifications.
The paper tackles the problem of measuring how well process data comply with declarative process specifications, which are defined using LTLf rules, by introducing a probabilistic framework that accounts for rule interplay, and it demonstrates applicability in real-world scenarios like discovery and drift detection.
Declarative process specifications define the behavior of processes by means of rules based on Linear Temporal Logic on Finite Traces (LTLf). In a mining context, these specifications are inferred from, and checked on, multi-sets of runs recorded by information systems (namely, event logs). To this end, being able to gauge the degree to which process data comply with a specification is key. However, existing mining and verification techniques analyze the rules in isolation, thereby disregarding their interplay. In this paper, we introduce a framework to devise probabilistic measures for declarative process specifications. Thereupon, we propose a technique that measures the degree of satisfaction of specifications over event logs. To assess our approach, we conduct an evaluation with real-world data, evidencing its applicability in discovery, checking, and drift detection contexts.