An Inductive Miner Implementation for the PM4Py Framework
BSc Thesis Project
Title: An Inductive Miner Implementation for the PM4Py Framework
Author: Timo Pohl
1st Examiner: Prof.dr.ir. Wil M.P. van der Aalst
2nd Examiner: Prof. Dr. rer. nat. Bernhard Rumpe
Daily Supervisor: Marco Pegoraro
Event logs record data (process traces) about the execution of business processes in companies and organisations. This data can be a valuable source of information to validate, improve and visualize the actual process executions. Process discovery, a sub discipline of process mining, aims at extracting so called process models out of the recorded process traces. These process models are used, for example, to visualize the process or to apply further, more elaborated process analysis techniques.
The goal of my bachelor thesis was to implement two process discovery algorithms described in the PhD thesis by Sander Leemans: the basic inductive miner and the infrequent inductive miner. In addition, the goal was to validate the implementations and to integrate them into the PM4Py framework.
To compare the quality of the process models generated by my implementations as well as their runtime behaviour, I designed and executed two experiments comparing my implementations to a process discovery algorithm already implemented in the PM4Py framework.