计算机应用

• 数据库技术(Database technology) • 上一篇    下一篇

基于同步管理器的过程挖掘算法

黄红梅 章云   

  1. 广东工业大学;广东水利电力职业技术学院 广东工业大学自动化学院
  • 收稿日期:2008-06-06 修回日期:2008-07-14 发布日期:2008-12-01 出版日期:2008-12-01
  • 通讯作者: 黄红梅

Process mining algorithm based on synchronized-manager

Hong-Mei Huang Yun ZHANG   

  • Received:2008-06-06 Revised:2008-07-14 Online:2008-12-01 Published:2008-12-01
  • Contact: Hong-Mei Huang

摘要: 强循环、弱循环、跳、不可见任务等复杂结构是业务流程中常见的流程结构,从事件日志中挖掘这些复杂结构是亟待解决的问题。分析了强循环、弱循环、跳、不可见任务的特征并给出了它们的判定定理,在此基础上结合同步管理器模型提出了γ+算法,该算法克服了目前过程挖掘算法的限制,在挖掘流程结构的同时挖掘管理行为,能够处理强循环、弱循环、跳、不可见任务等复杂结构,实例分析表明了算法的有效性与可适用性。

关键词: 过程挖掘, 强循环, 跳, 不可见任务, 同步管理器

Abstract: Strong-loop, weak-loop, jump and invisible tasks are ubiquitous between tasks in the business process, and mining them from the event logs is a problem to be solved. The features of strong-loop, weak-loop, jump and invisible tasks were analyzed, and the determinant theorems were offered, on the basis of which the γ+ algorithm was proposed with synchronized-manager models. The γ+algorithm overcomes the limitation of the converntional mining algorithms, executes the mining of management actions along with the mining of process structure,and manages to solve problems such as strong-loop, weak-loop, jump and invisible tasks. Experimental results indicate the effectiveness and applicability.

Key words: process mining, strong-loop, jump, invisible tasks, synchronized-manager