Discovery of Hybrid Process Models

Applying process mining to discover a hybrid Petri net and a causal graph from real-life data. The event log records business lawsuits from the Court of Justice of the State of Sao Paulo in Brazil.


This is a companion discussion topic for the original entry at https://kni.me/w/_jr8KszZj9Pj_c8G

Hi, thank you for uploading the workflow I want to interpret the ‚ÄúCausal Graph Miner‚ÄĚ

  1. The Root Node says 1, I guess is because it is 100% of the logs
  2. From the ‚ÄúDistribu√ɬ≠do por Depend√ɬ™ncia (movimenta‚Ķ‚ÄĚ we have a value of 041 does it means that 41% of the time the process goes to ‚ÄúRemetido ao DJE‚ÄĚ?
  3. From ‚ÄúRemetido ao DJE‚ÄĚ from ‚ÄúCertid√ɬ£o de Publica√ɬß√ɬ£o Expedida‚ÄĚ the value is 096, so I beg if you can tell me what is the interpretation for the 096

Thank you

Best Regards

Let’s tag the author the workflow, @PM4KNIME to see if they can help

1 Like

Dear @mauuuuu5 ,

in your screenshot, decimal numbers are not displayed correctly. Currently, I cannot reproduce the issue, but we will investigate it to fix it in future releases.

The metrics that can be selected in the legend panel are defined in the paper: http://sebastiaanvanzelst.com/wp-content/uploads/2022/08/Jul-15-Mining_for_Long_Term_Dependencies_in_Causal_Graphs-1.pdf

By default, the edges are annotated using the metrics Caus (defined in Definition 3) and LD (defined in Definition 5).

Kind regards

1 Like