工作流模型结构化挖掘方法研究  

Research of workflow model's structural mining method

作　　者：王海涛[1] 

机构地区：[1]山东大学计算机科学与技术学院,山东济南250022

出　　处：《计算机工程与设计》2007年第17期4223-4228,共6页Computer Engineering and Design

摘　　要：流程挖掘是业务流程管理(business process management,BPM)研究的一项重要内容。提出了一种结构化挖掘方法,实现从事务型日志中挖掘出工作流网。该方法基于工作流模型的4种基本结构(顺序、并行、选择和循环)进行挖掘。定义了可挖掘的工作流模型——结构化工作流网(structural workflow net,SWF),从日志预处理,流程挖掘方法和合理性验证3个方面对挖掘算法进行了详细描述,证明挖掘出的工作流模型满足合理性和安全性的同时,具有可读性和容易理解的特点。Process mining is one of the important contents of business process management. A structural mining method is presented to realize workflow mining from event logs. The method mines the workflow models based on the four kinds of basic structure ofworkflow （sequence, parallel, alternative and loop）. Firstly, a definition of what kinds of workflow nets can be rediscovery is given, that is structure workflow net （SWF）. After that, a detailed description of mining is presented including three sections： pre_process of log, arithmetic of process mining and reasonable validation of model. At last, the result proves that the workflow net mined using this method is reasonable, security, and more understandable.

关 键 词：流程 挖掘 事件日志 算法 工作流网 

分 类 号：TP302[自动化与计算机技术—计算机系统结构]