URL study guide
https://tue.osiris-student.nl/onderwijscatalogus/extern/cursus?cursuscode=2IMI00&collegejaar=2026&taal=enOmschrijving
Organizations are constantly trying to improve the way processes are performed including administrative processes, logistics processes, or knowledge-intensive processes. Typically, such operational processes are supported by information systems that record events as they take place in real life in so-called event logs. Process Mining techniques allow for extracting information from event logs to discover models describing processes, organizations, and products, detect deviations, or predict outcomes and performance of processes. Unlike classical data mining techniques the objective of process mining is to describe and predict a large number of features in a single model at once such as sequencing, choices, parallelism and loops of multiple activities, involvement of resources in activities, or the influence of data on the outcomes of activities.In this seminar, we will study and discuss the state-of-the-art research on Process Mining in the context of practically relevant uses cases from industry. We will discuss works along the entire process mining life-cycle: event data extraction and event log construction, various event data pre-processing methods, process model discovery, conformance checking and deviation detection, prediction based on models learned from event data, and visualization techniques. Additionally, students will be exposed to data mining and machine learning solutions that are relevant for process mining problems. This provides students a good insight into this field of research in preparation for a Master project in the PA group.
Students will study, present, and discuss selected research articles and book chapters, as well as Master theses. Through this material, students get an overview on the different types of problems faced in process mining from a practical perspective as well as the various research approaches to address them. Through presenting and discussing articles, students will learn to evaluate research approaches and evaluation techniques regarding the research problem and reproducibility. Through presenting and discussing previous Master theses, and by reproducing an evaluation from a scientific paper using process mining software and tools, students gain an understanding of “patterns” and “anti-patterns” in the execution of a Master project and in writing a thesis.
Students enrolling in the seminar are invited to propose their own relevant literature and topics in preparation for their Master project, or to pick from a broad list of available topics and literature made available before the start of the seminar.
Doelstellingen
This seminar combines teaching research methods (in preparation for a Master project) with providing students with recent and ongoing research in the area of event data analysis and process analysis. We study recent research articles, book chapters, and Master theses on topics along the entire analysis life-cycle. Through presentation and group discussions, we work out how to critically approach, execute, evaluate, and discuss research questions. The aim of the seminar is to prepare students for their graduation project. The specific literature and topics studied in the seminar will be made available before the first meeting for student seection; students may propose their own relevant literature and topics in preparation for their Master project. At the end of this seminar, students:• Have an overview on the state of the art in research (general research problems, techniques, and practically relevant use cases) in process mining and event data analysis
• Are able to independently study, present, discuss, compare, and criticize scientific texts
• Can discuss and select common research methods and techniques for a chosen problem
• Are able to propose and critically discuss research questions and experimental setups to test research questions
• Are able to conduct an experimental evaluation and discuss its validity regarding conclusions and reproducibility