[ProM-developers] Help for case studies for ProM

JBuijs

Vincenzo Madeo posted this message on the ProM developers mailing list on 3-11-2010 14:39 GMT+1 which might be interesting for you:

Dear
all,

I am a student and I am willing to use some analysis ProM plug-Ings (Process
Analysis and Performance) for my Master Thesis.



My problem is that I need of case studies as real benchmark
datasets and I tried to find them from the web, as usual in data mining
context, but unfortunately there is no way I could find them.

Could
you tell me where I can find (If there are) the case studies for ProM?

Thank
you in advance

Vincenzo Madeo

JBuijs

Boudewijn van Dongen replied:

Dear Vincenzo,

 

Of course, we cannot provide you with any real-life
datasets, as these have been provided to us under non-disclosure agreements.
However, it is our experience that it is very easy to obtain data from
companies, once they see the relevance of process mining for their business.
For the conversion of datasets to formats that can be read by ProM, I refer to
ProMimport and XESAme.

 

Also, if you want to make your own data using CPN tools,
then you can follow the tutorial on the ProM website (http://prom.win.tue.nl/research/wiki/tools/promcpn).
This tutorial shows how to use CPN tools to generate event logs. This allows
you to make logs as big, real-life, etc. as you want them to be.

 

Kind regards,

Boudewijn van Dongen

JBuijs

Andrea Burattin added the following:

Dear Vincenzo,

currently there is no "standard dataset" for
the benchmarking of process mining algorithms.

If you need to generate also the process (not only the
process logs, as with CPN tools) please take a look at PLG:

http://www.processmining.it/sw/plg.
It's still a young project and has some limitations, but can be useful as a
starting point.

 

Regards

 

--

Andrea Burattin, PhD student

JBuijs

Vincenzo replied:

Dear Boudewijin, Andrea and Janssen,

thanks a lot for your quickly answer and  your suggestions. 

Let me do a brief consideration, in my experience it is very hard to obtain

data from Italian companies for two main reasons (I assumed): 

1) privacy issue -- this is reasonable -- 2) but also because in my opinion

companies don’t spend their money in logging process capabilities.---On the

other hand it is very hard to convince these companies of the relevance of

process mining for their business if no case studies are available.

Btw, I would be very grateful if somebody could provide me of the contact of

some "available" companies that I can contact to obtain such real dataset.

Thanks again

Vincenzo Madeo