A great demand for upsampling event logs exists to carry out process-mining tasks in a more accurate form. For instance, when conducting predictive process monitoring analytics, we require more process data for analysis, enabling us to provide the best advice for decision-makers. Additionally, the generated event logs can address the issue of lacking training data due to privacy and security restrictions on business processes. This paper explores alternative methods for discovering event-log generative models. Experiments were conducted on both real and synthetic event logs. The results show that our techniques can generate event logs that are closer to the original datasets than what state-of-the-art techniques are capable of.
A great demand for upsampling event logs exists to carry out process-mining tasks in a more accurate form. For instance, when conducting predictive process monitoring analytics, we require more process data for analysis, enabling us to provide the best advice for decision-makers. Additionally, the generated event logs can address the issue of lacking training data due to privacy and security restrictions on business processes. This paper explores alternative methods for discovering event-log generative models. Experiments were conducted on both real and synthetic event logs. The results show that our techniques can generate event logs that are closer to the original datasets than what state-of-the-art techniques are capable of.
Techniques for Discovering Event-Log Generative Models
WU, JIANI
2022/2023
Abstract
A great demand for upsampling event logs exists to carry out process-mining tasks in a more accurate form. For instance, when conducting predictive process monitoring analytics, we require more process data for analysis, enabling us to provide the best advice for decision-makers. Additionally, the generated event logs can address the issue of lacking training data due to privacy and security restrictions on business processes. This paper explores alternative methods for discovering event-log generative models. Experiments were conducted on both real and synthetic event logs. The results show that our techniques can generate event logs that are closer to the original datasets than what state-of-the-art techniques are capable of.File | Dimensione | Formato | |
---|---|---|---|
Wu_Jiani_Thesis.pdf
accesso riservato
Dimensione
24.29 MB
Formato
Adobe PDF
|
24.29 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/52282