PLG2: Multiperspective Process Randomization with Online and Offline Simulations

The evaluation of process mining algorithms requires, as any other data mining task, the availability of large amount of (real-world) data. Despite the increasing availability of such datasets, they are affected by many limitations: in primis, the absence of a "gold standard" (i.e., the reference model). This work extends an approach already available in the literature for the generation of random processes. Novelties have been introduced throughout the work which, in particular, involve the complete support for multiperspective models and logs (i.e., the control-flow perspective is enriched with time and data information) and for online settings (i.e., generation of multiperspective event streams and concept drifts). The proposed new framework is able to cover the spectrum of possible scenarios that can be observed in the real-world.

More info: https://andrea.burattin.net/publications/2016-bpm-demo

Transcript

1. (Process) mining algorithms need gold standard for their validation (i.e., the ground truth) Process Mining Type Input Gold standard Control-flow discovery Event log Process model Conformance analysis Event log + Process model Process model (correct) Model enhancement Event log + Process model Process model
2. (Process) mining algorithms need gold standard for their validation (i.e., the ground truth) Process Mining Type Input Gold standard Control-flow discovery Event log Process model Conformance analysis Event log + Process model Process model (correct) Model enhancement Event log + Process model Process model Real datasets (e.g., BPI Challenges) only provides the input, no gold standard
3. (Process) mining algorithms need gold standard for their validation (i.e., the ground truth) Process Mining Type Input Gold standard Control-flow discovery Event log Process model Conformance analysis Event log + Process model Process model (correct) Model enhancement Event log + Process model Process model Real datasets (e.g., BPI Challenges) only provides the input, no gold standard Internal Process Representation Random Process Generator Random Process Modifier PLG2:
4. (Process) mining algorithms need gold standard for their validation (i.e., the ground truth) Process Mining Type Input Gold standard Control-flow discovery Event log Process model Conformance analysis Event log + Process model Process model (correct) Model enhancement Event log + Process model Process model Real datasets (e.g., BPI Challenges) only provides the input, no gold standard Internal Process Representation Process Importer (as BPMN) Process Exporter Random Process Generator Random Process Modifier PLG2:
5. (Process) mining algorithms need gold standard for their validation (i.e., the ground truth) Process Mining Type Input Gold standard Control-flow discovery Event log Process model Conformance analysis Event log + Process model Process model (correct) Model enhancement Event log + Process model Process model Real datasets (e.g., BPI Challenges) only provides the input, no gold standard Internal Process Representation Process Importer (as BPMN) Process Exporter Random Process Generator Random Process Modifier Random Noise Generator Event Log Generator Static File (XES or MXML) PLG2:
6. (Process) mining algorithms need gold standard for their validation (i.e., the ground truth) Process Mining Type Input Gold standard Control-flow discovery Event log Process model Conformance analysis Event log + Process model Process model (correct) Model enhancement Event log + Process model Process model Real datasets (e.g., BPI Challenges) only provides the input, no gold standard Internal Process Representation Process Importer (as BPMN) Process Exporter Random Process Generator Random Process Modifier Random Noise Generator Event Log Generator Event Stream Generator Static File (XES or MXML) TCP Network Stream PLG2:
7. (Process) mining algorithms need gold standard for their validation (i.e., the ground truth) Process Mining Type Input Gold standard Control-flow discovery Event log Process model Conformance analysis Event log + Process model Process model (correct) Model enhancement Event log + Process model Process model Real datasets (e.g., BPI Challenges) only provides the input, no gold standard Internal Process Representation Process Importer (as BPMN) Process Exporter Random Process Generator Random Process Modifier Random Noise Generator Event Log Generator Event Stream Generator Static File (XES or MXML) TCP Network Stream Fundamental features of PLG2 PLG2:
8. (Process) mining algorithms need gold standard for their validation (i.e., the ground truth) Process Mining Type Input Gold standard Control-flow discovery Event log Process model Conformance analysis Event log + Process model Process model (correct) Model enhancement Event log + Process model Process model Real datasets (e.g., BPI Challenges) only provides the input, no gold standard Internal Process Representation Process Importer (as BPMN) Process Exporter Random Process Generator Random Process Modifier Random Noise Generator Event Log Generator Event Stream Generator Static File (XES or MXML) TCP Network Stream Fundamental features of PLG2 Multi-perspective models PLG2:
9. (Process) mining algorithms need gold standard for their validation (i.e., the ground truth) Process Mining Type Input Gold standard Control-flow discovery Event log Process model Conformance analysis Event log + Process model Process model (correct) Model enhancement Event log + Process model Process model Real datasets (e.g., BPI Challenges) only provides the input, no gold standard Internal Process Representation Process Importer (as BPMN) Process Exporter Random Process Generator Random Process Modifier Random Noise Generator Event Log Generator Event Stream Generator Static File (XES or MXML) TCP Network Stream Fundamental features of PLG2 Python scripts for attributes Multi-perspective models PLG2:
10. (Process) mining algorithms need gold standard for their validation (i.e., the ground truth) Process Mining Type Input Gold standard Control-flow discovery Event log Process model Conformance analysis Event log + Process model Process model (correct) Model enhancement Event log + Process model Process model Real datasets (e.g., BPI Challenges) only provides the input, no gold standard Internal Process Representation Process Importer (as BPMN) Process Exporter Random Process Generator Random Process Modifier Random Noise Generator Event Log Generator Event Stream Generator Static File (XES or MXML) TCP Network Stream Fundamental features of PLG2 Python scripts for attributes Python scripts for times Multi-perspective models PLG2:
11. (Process) mining algorithms need gold standard for their validation (i.e., the ground truth) Process Mining Type Input Gold standard Control-flow discovery Event log Process model Conformance analysis Event log + Process model Process model (correct) Model enhancement Event log + Process model Process model Real datasets (e.g., BPI Challenges) only provides the input, no gold standard Internal Process Representation Process Importer (as BPMN) Process Exporter Random Process Generator Random Process Modifier Random Noise Generator Event Log Generator Event Stream Generator Static File (XES or MXML) TCP Network Stream Fundamental features of PLG2 Python scripts for attributes Python scripts for times Multi-perspective models On-the-fly switch stream source PLG2:
12. (Process) mining algorithms need gold standard for their validation (i.e., the ground truth) Process Mining Type Input Gold standard Control-flow discovery Event log Process model Conformance analysis Event log + Process model Process model (correct) Model enhancement Event log + Process model Process model Real datasets (e.g., BPI Challenges) only provides the input, no gold standard Internal Process Representation Process Importer (as BPMN) Process Exporter Random Process Generator Random Process Modifier Random Noise Generator Event Log Generator Event Stream Generator Static File (XES or MXML) TCP Network Stream Fundamental features of PLG2 Python scripts for attributes Python scripts for times Multi-perspective models On-the-fly switch stream source PLG2: PLG 2