Quantity of data

Hoping that this is not considered a technical question, I have the following problem: ProM 6.3 does not seem to be able to handle the large amount of data. I can have split up the 7 or so million records of the "logged-in" clicks in several groups (based on age category) and then further split in Male and Female, even removed consecutive clicks if they are within the same session ID on the same page. I can load the data into ProM (+40000 traces, +170000 events), but most of the modules - e.g. Inductive Miner, Simple Heuristics Filter, Dotted Chart all end up either crashing ProM or hanging. Is this due to lack of memory - I run ProM on OSX 10.6.8 (snow leopard) with 4GB of memory.


  • bfvdongebfvdonge Posts: 23
    As with many data analysis challenges, part of the challenge is the amount of data. On a computer with "only" 4 GB of memory, I can imagine that ProM has difficulties dealing with the full dataset. You could consider ways to limit the amount of data considered for the initial analysis. How to do so is part of the challenge!
  • AnneAnne Posts: 13

    Feel free to request a Disco (https://fluxicon.com/disco/) license via anne@fluxicon.com if you want to use Disco for the challenge. The data loads without problem. 

    Mobile: +31-(0)62-436-4201 Web: http://www.fluxicon.com Blog: http://fluxicon.com/blog/
Sign In or Register to comment.