Abstract: To solve the problem that the existing mining algorithm could not mine simplest 2-loops from local complete log traces without a fixed behavior pattern like “aba”,an αL -algorithm by extending classic α-algorithm was presented.The definition of simplest 2-loop was given by petri nets.The 2-loop and concurrency structure were distinguished according to the principle that different structures had different behaviors in event logs.After that,the same type of 2-loops structure was accurately divided by following degrees of tasks.The real data was used for extensive experiments sets to show the effectiveness and correctness of proposed method.In addition,the αL -algorithm was implemented in ProM tool.

Key words: business process management, process mining, following degrees, log completeness, Petri nets

摘要： 为解决现有挖掘算法无法从不具有显式行为特征“aba”的局部完备性的日志文件中挖掘最简2度循环的问题,对经典的alpha算法进行了扩展,提出了αL算法,用于从具有行为特征或不具有行为特征的局部完备性日志中挖掘出最简2度循环。给出了最简2度循环的形式化定义。从全局角度根据不同结构具有不同行为轨迹来区分最简2度循环与并发结构。通过紧邻关系的强弱,来有效识别并发分支上同类型的最简2度循环结构。通过大量模型实验表明了所提算法的有效性,并将算法集成在开源框架ProM中。

关键词: 业务流程管理, 流程挖掘, 紧邻度, 日志完备性, Petri网