.png){kind=icon}
返回
Significant stochastic dependencies in process models
.png){kind=icon}
PRE
AI
DOI：10.1016/j.is.2023.102223.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Process mining aims to artificially derive meaningful information from event logs recorded from information systems that support business processes in organisations. In many real-life processes, decisions do not only depend on the current state of the model, but also on decisions made earlier in the process. Such dependencies challenge process mining techniques: Petri nets are limited to hard dependencies (non-free-choice constructs). In this paper, we study stochastic dependencies, which model that the likelihood of decisions in a process may change dynamically based on earlier decisions. We introduce a modelling formalism that supports stochastic dependencies by extending stochastic labelled Petri nets, we study symmetries of this formalism, introduce a stochastic process discovery technique that discovers such models and we adapt two conformance checking techniques to validate the discovered models. The techniques have been implemented, and evaluated on computational feasibility and applicability. Finally, we show that the quality of our new models can compete with existing discovery techniques, and we show that stochastic dependencies are present in existing real-life logs and may lead to new types of insights.& COPY; 2023 Published by Elsevier Ltd.
Keyword:
Process mining
Stochastic process mining
Stochastic process discovery
Model dependencies
Long-distance dependencies
期刊
IF:
3.9
论文数:
2.8K
被引数:
1.8K
机构
引用论文
.png){kind=icon}
Comfort-Enhanced Daily Disposable Contact Lens Reduces Symptoms Among Weekly/Monthly Wear Patients
PREAI
分享收藏
查看更多内容