In this episode of A Side Of Data, I interview Sebastiaan van Zelst about his experience of building a pm4py python library for process mining.

The show timeline:

1:30 How did pm4py start? The reasons behind

5:30 Why do you need a new library? Why python?

8:00 How big is the library now? What is the current progress

12:00 Categorizing and comparing process mining tools: programming libraries, strand-alone academic, commercial (Apromore, Prom, pm4py, Celonis, Fluxicon Disco, Minit, Process Gold, etc.)

17:30 Problems with commercial tools

20:30 How challenging was to build the pm4py. How easy is it to use the library with the coding workflow on python.

25:40 Projects built on top of the pm4py

29:00 What is available in the library now

32:00 How to start with pm4py and with process mining

34:20 About the PM4Knime project.

37:00 extra: What is the interactive process discovery, how pm4py supports it.

Additional info:

The library: http://pm4py.org 

The paper on incremental discovery: https://sebastiaanvanzelst.com/wp-content/uploads/2020/03/Incremental_Discovery_of_Hierarchical_Process_Models__RCIS_1.pdf

Interviewee page: https://sebastiaanvanzelst.com/?page_id=62 

Mentioned telegram bot: https://github.com/delas/pmbot

Find A SIDE OF DATA on twitter: https://twitter.com/ASideOFData

And on the web: asideofdata.com

SEND us an email with suggestions: [email protected]

Listen to more A Side Of Data:

Building a process mining startup with Raffaele Conforti: https://soundcloud.com/asideofdata/building-a-process-mining-startup-with-raffaele-conforti

New decade for process mining with Wil van der Aalst: https://soundcloud.com/asideofdata/new-decade-for-process-mining-with-wil-van-der-aalst