Measuring the interestingness of temporal logic behavioral specifications in process mining

https://doi.org/10.1016/j.is.2021.101920Get rights and content
Under a Creative Commons license
open access

Highlights

  • Exploitation of declarative linear-temporal-logic rules shaped as if-A-then-B rules.

  • Association rule mining interestingness measures applied to temporal logic formulae.

  • Measures build upon a probabilistic interpretation of linear temporal logic formulae.

  • Computing the measures requires linear time and space in the size of the input data.

  • Measures have different reactions in presence of noise depending on the given rule.

Abstract

The assessment of behavioral rules with respect to a given dataset is key in several research areas, including declarative process mining, association rule mining, and specification mining. An assessment is required to check how well a set of discovered rules describes the input data, and to determine to what extent data complies with predefined rules. Particularly in declarative process mining, Support and Confidence are used most often, yet they are reportedly unable to provide a sufficiently rich feedback to users and cause rules representing coincidental behavior to be deemed as representative for the event logs. In addition, these measures are designed to work on a predefined set of rules, thus lacking generality and extensibility. In this paper, we address this research gap by developing a measurement framework for temporal rules based on (LTLpf). The framework is suitable for any temporal rules expressed in a reactive form and for custom measures based on the probabilistic interpretation of such rules. We show that our framework can seamlessly adapt well-known measures of the association rule mining field to declarative process mining. Also, we test our software prototype implementing the framework on synthetic and real-world data, and investigate the properties characterizing those measures in the context of process analysis.

Keywords

Declarative process mining
Specification mining
Association rule mining
Interestingness measures
Temporal rules

Cited by (0)

1

Authors in alphabetical order.