以Artifact为中心的业务流程扩展二部图模型及其相似性度量方法  

Extended bipartite graph model and similarity matching method for Artifact-centric business process models

作　　者：刘海滨[1] 刘国华[2] 方巍[1] 柴朝华[1] 谢会芹[1] 

机构地区：[1]河北科技师范学院工商管理学院,河北秦皇岛066004 [2]东华大学计算机科学与技术学院,上海201620

出　　处：《计算机集成制造系统》2017年第5期1050-1059,共10页Computer Integrated Manufacturing Systems

基　　金：国家社会科学青年基金资助项目(16CGJ007);河北省自然科学青年基金资助项目(F2015407039);河北省社会科学基金资助项目(HB15GL073);河北省社会科学发展研究课题民生调研专项资助项目(201601138)~~

摘　　要：为了保证计算结果的准确性和有效性,提出一种以Artifact为中心的业务流程扩展二部图模型及其相似性度量方法。首先提出一个用于描述以Artifact为中心的业务流程的扩展二部图模型,将流程模型匹配转换为图匹配问题;其次提出一种利用矩阵转换操作来度量图编辑距离的方法;在扩展二部图模型的基础上结合属性划分技术,分别提出服务节点和仓库节点行为相似度计算方法;设计实现了以Artifact为中心的业务流程扩展二部图的图匹配算法。理论分析和实验结果表明,该算法在查询时间和执行效率上优于现有算法,具有较高的查准率和检索质量。To guarantee the accuracy and effectiveness of computational result, an extended bipartite graph model of Artifact-centric business process was presented, which could transform process model matching to graph matching. A new method by using matrix conversion to measure graph edit distance was also brought forward. According to the attribute partition technology, the similarity measurement of service and repertory were proposed respectively. A graph matching algorithm of an extended bipartite graph supporting artifact-centric business processes was designed. Theoretical analysis and experimental results showed that the presented algorithm was superior to the exist- ing algorithms in a number of performance metrics such as execution time and precision.

关 键 词：业务流程管理 流程相似性 图匹配 ARTIFACT 属性划分 

分 类 号：TP311.52[自动化与计算机技术—计算机软件与理论]