Dealing with Complex Parallel Structures in Process Discovery
Name
Bogdan Semiletko
Abstract
One of the aims of process mining is to discover a process model from a log. However, the quality of the discovered model depends on the completeness of the information about the process behaviour contained in the log. Incomplete logs do not provide all the possible behaviours. Existing process discovery algorithms dealing with incomplete logs, have troubles when working with complex parallel structures, because parallel behaviour has factorial rate of growth with respect to the number of branches. In this work, a new algorithm is proposed, which combines divide and conquer approach, with the existing mining algorithms to improve discovery of highly structured and highly concurrent process models from incomplete logs. This work describes the proposed algorithm, and explains how it works with illustrative step-by-step examples of the mining procedure. Finally, we evaluate the effectiveness and efficiency of our approach by using process models containing complex parallel structures and randomly generated models.
Graduation Thesis language
English
Graduation Thesis type
Master - Software Engineering
Supervisor(s)
Fabrizio Maria Maggi
Defence year
2015