In Process Mining, computing alignments is a conformancechecking technique to compare a process model with an event log of the same process to pinpoint difference between how the model would prescribe the process to be executed, and how the event log states the process has been executed. The complexity of this problem is naturally exponential with respect to the size of the model, and benefits can be achieved using divideandconquer approaches: the model is decomposed into small fragment for which we can compute alignments. This thesis compares the time to compute alignments using the traditional approaches and our decompositionbased approaches to identify the possible benefits. The results are also compared with different approaches based on processmodel decompositions.
In Process Mining, computing alignments is a conformancechecking technique to compare a process model with an event log of the same process to pinpoint difference between how the model would prescribe the process to be executed, and how the event log states the process has been executed. The complexity of this problem is naturally exponential with respect to the size of the model, and benefits can be achieved using divideandconquer approaches: the model is decomposed into small fragment for which we can compute alignments. This thesis compares the time to compute alignments using the traditional approaches and our decompositionbased approaches to identify the possible benefits. The results are also compared with different approaches based on processmodel decompositions.
Greedy Approach to Compute Alignments of Process Models and Event Logs
CHIARELLO, SOFIA
2022/2023
Abstract
In Process Mining, computing alignments is a conformancechecking technique to compare a process model with an event log of the same process to pinpoint difference between how the model would prescribe the process to be executed, and how the event log states the process has been executed. The complexity of this problem is naturally exponential with respect to the size of the model, and benefits can be achieved using divideandconquer approaches: the model is decomposed into small fragment for which we can compute alignments. This thesis compares the time to compute alignments using the traditional approaches and our decompositionbased approaches to identify the possible benefits. The results are also compared with different approaches based on processmodel decompositions.File  Dimensione  Formato  

Chiarello_Sofia.pdf
accesso aperto
Dimensione
7.11 MB
Formato
Adobe PDF

7.11 MB  Adobe PDF  Visualizza/Apri 
The text of this website © Università degli studi di Padova. Full Text are published under a nonexclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/61404