Data-driven Process Discovery - Artificial Event Log
datasetposted on 08.12.2016 by Felix Mannhardt
Datasets usually provide raw data for analysis. This raw data often comes in spreadsheet form, but can be any collection of data, on which analysis can be performed.
A synthetic event log with 100,000 traces and 900,000 events that was generated by simulating a simple artificial process model. There are three data attributes in the event log: Priority, Nurse, and Type. Some paths in the model are recorded infrequently based on the value of these attributes. Noise is added by randomly adding one additional event to an increasing number of traces. CPN Tools (http://cpntools.org) was used to generate the event log and inject the noise.