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.

Preformance issues and general help (can send data)

We have just begun to use ProM, and in using it we have come across a couple of questions. Our data is originally in .csv format so we convert it to a .xes and change the XFactor to be an XESlite format.

1. We were wondering with our data if ProM6.7 was able to handle this much data? (There are 4097 traces and 244000 events) Is there a limit to how many data points ProM6.7 can handle at once and if so do we exceed this limit? This data set is only a very small portion of what we intend to put into ProM6.7. There has been a lot of waiting for ProM to return different nets and it normally takes too long to run the visuals so we end up quitting out of them. Each time we quit an event we must restart our computer and don't think that this is supposed to be happening.
We have 16GB of Ram and Intel Core i7 vPro 7th gen processor.

2. We have tried several different plug-ins, none of which have illustrated meaningful process flows that we have been able to visualize. Though we know our data is not that algorithmic, we were wondering if there are other plugins that can provide us with meaningful insights for our data.  We are more than willing to send you are data if you have extra time to look over it. We would love to know if we have our data formatted properly and if there is an easier way to split our data to still provide accurate results.

Our data follows this format.
6 columns(Start Time, End Time, Invoice Number, Reason Code, Action Code, Disposition Code) 122000 rows (observations)

We currently have it set up so that our traces are the Invoice Number and our events are the different codes.

Any help would be appreciated. Thanks.


Tagged:

Comments

  • kfgz087kfgz087 Posts: 61

    Dear johnsum2010,

    (1) You can set the memory size in the ProM Package Manager. The default setting is 4 GB. You can set the possible maximal value based on your hardware and check if it is available to import your large data.

    (2) The popular model discovery plugin is Inductive Visual Miner. The discovered insights of your data depend on how you map your data to the event log. In general, we consider the elements with time as events. Of course, you can consider other things as events if needed.

    You can send me (g.li.3@tue.nl) your data and I try to find some time to look at it.

    Guangming Li
    PhD student in Process Mining at Eindhoven University of Technology
Sign In or Register to comment.