Alignment-based conformance checking over probabilistic events

TL;DR

This paper extends alignment-based conformance checking to probabilistic event logs, enabling analysis of noisy, uncertain data from modern sensors and AI technologies against process models.

Contribution

It introduces a weighted trace model and alignment cost function, allowing conformance checking under uncertainty with a confidence threshold.

Findings

The algorithm effectively handles noisy event data.

It improves conformance detection in uncertain environments.

Demonstrated on real-life datasets.

Abstract

Conformance checking techniques allow us to evaluate how well some exhibited behaviour, represented by a trace of monitored events, conforms to a specified process model. Modern monitoring and activity recognition technologies, such as those relying on sensors, the IoT, statistics and AI, can produce a wealth of relevant event data. However, this data is typically characterised by noise and uncertainty, in contrast to the assumption of a deterministic event log required by conformance checking algorithms. In this paper, we extend alignment-based conformance checking to function under a probabilistic event log. We introduce a weighted trace model and weighted alignment cost function, and a custom threshold parameter that controls the level of confidence on the event data vs. the process model. The resulting algorithm considers activities of lower but sufficiently high probability that better align with the process model. We explain the algorithm and its motivation both from formal and intuitive perspectives, and demonstrate its functionality in comparison with deterministic alignment using real-life datasets.