University of Twente Student Theses

Login
As of Friday, 8 August 2025, the current Student Theses repository is no longer available for thesis uploads. A new Student Theses repository will be available starting Friday, 15 August 2025.

Comparative Study of Trace Clustering and Process Cube Sequences in Process Mining

Park, Byeonghun (2025) Comparative Study of Trace Clustering and Process Cube Sequences in Process Mining.

[img] PDF
1MB
Abstract:One of the key challenges that process mining often encounters is “spaghetti-like” process models generated in process discovery due to complex variants of the event data. To address this, segmentation techniques such as trace clustering and Process Cube analysis have been proposed. This study presents a comparative benchmark of these two approaches, both individually and in combination, applied to the 4TU Sepsis Cases event log. By segmenting the data using trace clustering, Process Cube, Process Cube followed by trace clustering (Process Cube → trace clustering), and trace clustering followed by Process Cube (trace clustering → Process Cube), the resulting process models are evaluated through complexity metrics such as Cyclomatic Number, Coefficient of Network Connectivity, and arc density. The results show that the Process Cube → trace clustering combination produces the most simplified and interpretable process models. The result also suggests that dimension-based segmentation provides a more effective segmentation for further clustering. This work focuses on the segmentation order and provides a methodological basis for applying appropriate techniques to improve model quality in process discovery.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:58 process technology
Programme:Business & IT BSc (56066)
Link to this item:https://purl.utwente.nl/essays/107445
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page