To prevent spam users, you can only post on this forum after registration, which is by invitation. If you want to post on the forum, please send me a mail (h DOT m DOT w DOT verbeek AT tue DOT nl) and I'll send you an invitation in return for an account.

Duplicate tasks and simplicity of a process model

This discussion was created from comments split from: Predict changes in ProM.
Joos Buijs

Senior Data Scientist and process mining expert at APG (Dutch pension fund executor).
Previously Assistant Professor in Process Mining at Eindhoven University of Technology

Comments

  • hi

    I read your thesis.  but I can not determines that process model should express duplicate task or don't allow to make duplicate task. because if process model have  alternative duplicate task cause to low simplicity.

    Thank you for helping

  • Dear Janan,

    I don't understand your question, but the main idea is that bigger process models are more complex. To determine the minimal size of a process model we can use the fact that each activity should be present once, and at least once.

    Note that the simplicity metric that I use in my thesis, 'useless nodes' works slightly different.

    If you have further questions please ask.
    Joos Buijs

    Senior Data Scientist and process mining expert at APG (Dutch pension fund executor).
    Previously Assistant Professor in Process Mining at Eindhoven University of Technology
  • thank you for replay. my means is control flow algorithm should be able to tackle some common structure (Representational bias: invisible action, duplicate task, non free choice,...). but duplicate task issue make me confuse. I don't know in my algorithm I have to remove all duplicate tasks in the event log or my algorithm should be able present duplicate task?
  • Dear Janan,

    Whether you want your algorithm to be able to handle duplicate tasks is up to you and what preference you give to each quality dimension. Note that although duplicate tasks can result in more complex process models, the discovered process models could also score better on replay and precision... It's always a trade-off, as I extensively discuss in my PhD thesis.
    Joos Buijs

    Senior Data Scientist and process mining expert at APG (Dutch pension fund executor).
    Previously Assistant Professor in Process Mining at Eindhoven University of Technology
  • Thank you for helping.


  • Dear Buijs

    I have a problem in And Split.

    I use And Split gateway to tackle with duplicate task. I have sent a file to beter understand.  for instance, C is And split between G and D and output of both go to H. I have another sequence C, H,D  in the sequence of my graph.  I want to create another And Split between created And Split and sequence. my question  can I use another And split between And split C,G,D and C, D, H. it means D is same between two And Split. 

    Thank



Sign In or Register to comment.