工作流网频繁子网挖掘研究进展  

Research progress of frequent subnets mining on workflow nets

作　　者：张书涵[1,3] 费超群 黄锡昆 李阳阳 ZHANG Shuhan;FEI Chaoqun;HUANG Xikun;LI Yangyang(Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences,Institute of Computing Technology,Beijing 100190;Academy of Mathematics and Systems Science Key Laboratory of MADIS,Chinese Academy of Sciences,Beijing 100190;University of Chinese Academy of Sciences,Beijing 100049)

机构地区：[1]中国科学院计算技术研究所智能信息处理重点实验室,北京100190 [2]中国科学院数学与系统科学研究院管理、决策与信息系统重点实验室,北京100190 [3]中国科学院大学,北京100049

出　　处：《高技术通讯》2022年第8期811-824,共14页Chinese High Technology Letters

基　　金：国家重点研发计划(2016YFB1000902);国家自然科学基金(61232015,21472412,61621003);中国博士后科学基金(2020TQ0341)资助项目。

摘　　要：本文总结了工作流网频繁子网挖掘的主流研究方向,包括从一维的日志进程中构造工作流网及其子网和从二维工作流网中挖掘其频繁子网结构,总结了其中的代表性方法及其优缺点。工作流网具有复杂、异构拓扑结构和完备性语义的特性,本文详细分析了将频繁模式挖掘(FPM)算法直接用于工作流网频繁子网挖掘存在的问题及缺陷。并给出了工作流网频繁子网挖掘的典型应用,包括异常检测、跨组织变体分析等。最后讨论了工作流网频繁子网挖掘中的研究难点和未来研究趋势。This paper summaries two main directions of frequent subnets mining on workflow nets, including constructing workflow nets and subnets from 1-dimention log processes data, and mining sub-structures from 2-dimention workflow nets. The advantages and disadvantages of these approaches are discussed. Due to the complex networked data structures and complete semantics that are contained in workflow nets, the issues of applying frequent pattern mining algorithms on workflow net mining are thoroughly analyzed. Furthermore, some typical applications of mining frequent subnets on workflow nets are presented, including anomalous pattern detection, cross-organizational variability analysis, and so on. Finally, some challenges and potential future directions concerning frequent subnets mining on workflow net are discussed.

关 键 词：频繁模式挖掘(FPM) 工作流网 子网挖掘 PETRI网 进程挖掘 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论]