An Algorithm for Mining Indirect Dependencies From Loop-Choice-Driven Loop Structure via Petri Nets
提出AlphaID算法，从事件日志中挖掘循环选择驱动循环结构中的间接依赖关系，扩展Petri网并定义新变迁触发规则，通过人工和真实案例验证有效性。
Process mining is an emerging technology used to extract, detect, and improve actual processes by extracting knowledge from event logs generated from information systems. In the production process, we can obtain the optimal process based on practical experience. Indirect dependencies may exist among different structures in the optimal process model discovered from the event log of the executions that perform better. However, the existing process mining algorithms cannot effectively mine the indirect dependencies among different structures. To compensate for this deficiency, an algorithm named <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AlphaID</i> is proposed in this article, and it can mine the indirect dependencies in a loop-choice-driven loop structure. First, two algorithms are proposed to efficiently identify loop sequences and choice sequences from event logs. Then, the concept of association rules is proposed to describe indirect dependencies among different structures. Next, we expand the ordinary Petri net and redefine the new transition firing rules to represent the process model obtained by <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AlphaID</i> . Finally, the correctness and effectiveness of the algorithm are verified by an artificial case and a real case. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">AlphaID</i> is integrated into the ProM which is an open-source process mining tool platform as a plug-in.