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.

Quality metrics

TalionTalion Posts: 15
Hi everybody,
I want to compare the models obtained with the heuristic miner and inductive miner by means of the 4 quality metrics: Fitness, Precision, Generalization and Simplicity. Is there a plugin in ProM 6.7 that can calculate these 4 metrics all together? Thank you in advance.


Comments

  • kfgz087kfgz087 Posts: 61
    Hi Talion,

    As far as I know, there is no plugins which can calculate these 4 metrics all together.
    You have to calculate them seperately. More precisely, you can compute fitness with the plugin "Replay a Log on Petri Net for Conformance Analysis" and compute precision and generalization with the plugin "Measure Precision/Generalization".

    Guangming Li
    PhD student in Process Mining at Eindhoven University of Technology
  • TalionTalion Posts: 15
    Ok, thank you kfgz087! Instead for simplicity what I have to do? I read that this metric can be calculated manually, which is the formula for computing the simplicity?
  • Hi,I have the same questions as you.how can we get the simplicity?
  • JBuijsJBuijs Posts: 910
    There are several ways to measure simplicity (/readibility/complexity) of a process model. There is no agreed measure right now. I suggest to research the literature and pick your favorite.

    There is, to my knowledge, also no implementation in ProM 6, there are in Prom 5.
    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
  • TalionTalion Posts: 15
    Ok, thank you! However, I cannot find on the web formulas for measuring simplicity but only definitions. Can you link me some papers showing how to measure simplicity?
  • JBuijsJBuijs Posts: 910
    I can't check but this literature review seems like a good start

    http://www.sciencedirect.com/science/article/pii/S0920548916302276

    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
  • TalionTalion Posts: 15
    Perfect. Thank you!
  • Hello everybody!!
    i have a same problem like Talion; but i wanted to make the comparion between apha, alpha+ and alpha ++ algorithme using different logs files. the procedure for calculating these metrics. can i use the same method give by kfgz087 ?

  • JBuijsJBuijs Posts: 910
    I would recommend to do so indeed.
    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
  • thanks sir Joos Buijs! but I have a problem with "Measure Precision/Generalization" plugins. i can't use it beacause it is not with a green color like another plusgins which are available. Please How could do? thanks in advance for the answer!!
  • hello! please i have another question.
    is it possible to compute precision and generalization quality metrics using behavioral appropriateness and structural appropriateness?
    I'm not understanding these notion "behavioral appropriateness and structural appropriateness" very well
Sign In or Register to comment.