Humboldt-Universität zu Berlin - Faculty of Mathematics and Natural Sciences - Process-Driven Architectures

Process Mining (VL/UE)

Prof. Dr. Matthias Weidlich

 

Content

One emerging branch of data science is process mining. In the field of process automation, process mining aims at deriving qualitative and quantitative insights on the execution of a process based on recorded events logs.

 

Structure

In the first part of the course, lectures and recitations will focus on the formal foundations and basic techniques of process mining. That includes algorithms for process discovery (constructing models from event data), conformance checking (identifying deviations between models and event data), and model extension (enriching models based on event data). The recitations will include a tutorial in which the students are exposed to real-world data and process mining tools.

The second part of the course will be organised as a seminar. Each student will be asked to read a recent research paper on process mining (selection from a given list) and give a critical assessment of the approach presented in the paper in the form of a 45min presentation.

The course will be given in English. The first lecture will take place on Thursday, 19th October, 2017.

 

Exam

There will be an oral exam at the end of the semester. To be eligible to take the exam, each student will be required to give a presentation (45min) on a research paper in the second part of the course. 

Exams will take place at the following dates:

  • 22.02.2018
  • 23.02.2018
  • 27.04.2018

For each date, there are multiple slots available (some time between 9:00 and 14:45). You can start to register on January 25, 2018. You need to register until 2 weeks before the date of the oral exam.  To register, you will need the respective form (please fill before printing) and also bring your student ID as well as your ID card. 

 

Credit Points

The course counts for 5 LP and is open for: Informatik, Master of Science (M.Sc.) Informatik, Master of Education (M.Ed.) Wirtschaftsinformatik, Master of Science (M.Sc.). The related area of specialisation is "Daten- und Wissensmanagement".

 

Dates

VL Do 9-11 RUD 26, Raum 0'313
UE Do 11-13 RUD 26, Raum 0'313

 

 

Lecture Slides 

Recitations

  • Assignment Sheet 1 - Behavioural Formalisms
  • Assignment Sheet 2 - Logs (Jupyter notebook)
  • Assignment Sheet 3 - Alpha-Miner
  • Assignment Sheet 4 - ProM Mining
  • Assignment Sheet 5 - Heuristics Miner
  • Assignment Sheet 6 - Inductive Miner
  • Assignment Sheet 7 - Measures
  • Assignment Sheet 8 - Conformance
  • Assignment Sheet 9 - Alignment
  • Assignment Sheet 10 - Enhancement 

     

    Presentation Slots

    Date Topic and Presenter
       

    November 16, 2017

    1) Event Abstractions 
    Murat Gökhan Yigit

     

    3) Interaction Mining
    Alexander Heemann
       

    December 14, 2017

    7) Stream-based Discovery
    Simon Hansen

     

    8) Discovery of BPMN Models
    Severin Hußmann

     

     

    January 18, 2018

    12) Decomposing Conformance Checking
    Franko Maximilian Hölzig
      13) Approximating Alignments
    Steffen Przybylowicz
       

    February 1, 2018

    14) Discovery of Roles
    Elias Baumann

     

    15) Temporal Anomaly Detection
    Aydin Sader

     

    16) Predictive Monitoring
    Ninos Yonan


    Presentation Topics 

    Note on the dates: topics 1-9 are likely to be scheduled for Nov 16 or Dec 14, 2017. The remaining topics will be presented in Jan and Feb 2018.

    1) Event Abstractions
    Thomas Baier, Jan Mendling, Mathias Weske: Bridging abstraction layers in process mining. Inf. Syst. (IS) 46:123-139 (2014) paper link
    2) Pattern-based Abstractions
    Felix Mannhardt, Massimiliano de Leoni, Hajo A. Reijers, Wil M. P. van der Aalst, Pieter J. Toussaint: From Low-Level Events to Activities - A Pattern-Based Approach. BPM 2016: 125-141 paper link
    3) Interaction Mining
    Arik Senderovich, Andreas Rogge-Solti, Avigdor Gal, Jan Mendling, Avishai Mandelbaum: The ROAD from Sensor Data to Process Instances via Interaction Mining. CAiSE 2016: 257-273 paper link
    4) Overlapping Log Entries
    Ahmed Awad, Nesma M. Zaki, Chiara Di Francescomarino: Analyzing and repairing overlapping work items in process logs. Information & Software Technology 80: 110-123 (2016) paper link
    5) Genetic Mining
    Wil M. P. van der Aalst, Ana Karla A. de Medeiros, A. J. M. M. Weijters: Genetic Process Mining. ICATPN 2005:48-69 paper link
    6) Discovery based on Regions
    Josep Carmona, Jordi Cortadella, Michael Kishinevsky, Alex Kondratyev, Luciano Lavagno, Alexandre Yakovlev: A Symbolic Algorithm for the Synthesis of Bounded Petri Nets. Petri Nets 2008: 92-111 paper link
    7) Stream-based Discovery
    Andrea Burattin, Alessandro Sperduti, Wil M. P. van der Aalst: Control-flow discovery from event streams. IEEE Congress on Evolutionary Computation 2014: 2420-2427 paper link
    8) Discovery of BPMN Models
    Raffaele Conforti, Marlon Dumas, Luciano García-Bañuelos, Marcello La Rosa: Beyond Tasks and Gateways: Discovering BPMN Models with Subprocesses, Boundary Events and Activity Markers. BPM 2014:101-117 paper link
    9) Handling Duplicated Tasks 
    Xixi Lu, Dirk Fahland, Frank J. H. M. van den Biggelaar, Wil M. P. van der Aalst: Handling Duplicated Tasks in Process Discovery by Refining Event Labels. BPM 2016: 90-107 paper link
    10) Alternative Measures for Model Precision
    Jorge Munoz-Gama, Josep Carmona: A Fresh Look at Precision in Process Conformance. BPM 2010:211-226 paper link
    11) Negative Events in Model Evaluation
    Seppe K. L. M. vanden Broucke, Jochen De Weerdt, Jan Vanthienen, Bart Baesens: Determining Process Model Precision and Generalization with Weighted Artificial Negative Events. IEEE Trans. Knowl. Data Eng. 26(8): 1877-1889 (2014) paper link
    12) Decomposing Conformance Checking
    Jorge Munoz-Gama, Josep Carmona, Wil M. P. van der Aalst: Single-Entry Single-Exit decomposed conformance checking. Inf. Syst. 46: 102-122 (2014) paper link
    13) Approximating Alignments
    Boudewijn F. van Dongen, Josep Carmona, Thomas Chatain, Farbod Taymouri: Aligning Modeled and Observed Behavior: A Compromise Between Computation Complexity and Quality. CAiSE 2017: 94-109 paper link
    14) Discovery of Roles
    Andrea Burattin, Alessandro Sperduti, Marco Veluscek: Business models enhancement through discovery of roles. CIDM 2013:103-110 paper link
    15) Temporal Anomaly Detection
    Andreas Rogge-Solti, Gjergji Kasneci: Temporal Anomaly Detection in Business Processes. BPM 2014:234-249 paper link
    16) Predictive Monitoring
    Fabrizio Maria Maggi, Chiara Di Francescomarino, Marlon Dumas, Chiara Ghidini: Predictive Monitoring of Business Processes. CAiSE 2014:457-472 paper link
    17) Queue Mining
    Arik Senderovich, Matthias Weidlich, Avigdor Gal, Avishai Mandelbaum: Queue mining for delay prediction in multi-class service processes. Inf. Syst. (IS) 53:278-295 (2015) paper link

     

    See AGNES for further details: