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.
Measuring the interestingness of temporal logic behavioral specifications in process mining / Cecconi, Alessio; De Giacomo, Giuseppe; Di Ciccio, Claudio; Maggi, Fabrizio Maria; Mendling, Jan. - In: INFORMATION SYSTEMS. - ISSN 0306-4379. - 107:(2022). [10.1016/j.is.2021.101920]
Measuring the interestingness of temporal logic behavioral specifications in process mining
De Giacomo, Giuseppe;Di Ciccio, Claudio;Maggi, Fabrizio Maria;
2022
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.File | Dimensione | Formato | |
---|---|---|---|
Cecconi_postprint_Measuring_2022.pdf
accesso aperto
Note: https://doi.org/10.1016/j.is.2021.101920
Tipologia:
Documento in Post-print (versione successiva alla peer review e accettata per la pubblicazione)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
740.35 kB
Formato
Adobe PDF
|
740.35 kB | Adobe PDF | |
Cecconi_Measuring_2022.pdf
solo gestori archivio
Tipologia:
Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza:
Tutti i diritti riservati (All rights reserved)
Dimensione
998.29 kB
Formato
Adobe PDF
|
998.29 kB | Adobe PDF | Contatta l'autore |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.